nips nips2005 nips2005-194 knowledge-graph by maker-knowledge-mining
Source: pdf
Author: Michael Shettel, Shaun Vecera, Michael C. Mozer
Abstract: Theories of visual attention commonly posit that early parallel processes extract conspicuous features such as color contrast and motion from the visual field. These features are then combined into a saliency map, and attention is directed to the most salient regions first. Top-down attentional control is achieved by modulating the contribution of different feature types to the saliency map. A key source of data concerning attentional control comes from behavioral studies in which the effect of recent experience is examined as individuals repeatedly perform a perceptual discrimination task (e.g., “what shape is the odd-colored object?”). The robust finding is that repetition of features of recent trials (e.g., target color) facilitates performance. We view this facilitation as an adaptation to the statistical structure of the environment. We propose a probabilistic model of the environment that is updated after each trial. Under the assumption that attentional control operates so as to make performance more efficient for more likely environmental states, we obtain parsimonious explanations for data from four different experiments. Further, our model provides a rational explanation for why the influence of past experience on attentional control is short lived. 1 INTRODUCTION The brain does not have the computational capacity to fully process the massive quantity of information provided by the eyes. Selective attention operates to filter the spatiotemporal stream to a manageable quantity. Key to understanding the nature of attention is discovering the algorithm governing selection, i.e., understanding what information will be selected and what will be suppressed. Selection is influenced by attributes of the spatiotemporal stream, often referred to as bottom-up contributions to attention. For example, attention is drawn to abrupt onsets, motion, and regions of high contrast in brightness and color. Most theories of attention posit that some visual information processing is performed preattentively and in parallel across the visual field. This processing extracts primitive visual features such as color and motion, which provide the bottom-up cues for attentional guidance. However, attention is not driven willy nilly by these cues. The deployment of attention can be modulated by task instructions, current goals, and domain knowledge, collectively referred to as top-down contributions to attention. How do bottom-up and top-down contributions to attention interact? Most psychologically and neurobiologically motivated models propose a very similar architecture in which information from bottom-up and top-down sources combines in a saliency (or activation) map (e.g., Itti et al., 1998; Koch & Ullman, 1985; Mozer, 1991; Wolfe, 1994). The saliency map indicates, for each location in the visual field, the relative importance of that location. Attention is drawn to the most salient locations first. Figure 1 sketches the basic architecture that incorporates bottom-up and top-down contributions to the saliency map. The visual image is analyzed to extract maps of primitive features such as color and orientation. Associated with each location in a map is a scalar visual image horizontal primitive feature maps vertical green top-down gains red saliency map FIGURE 1. An attentional saliency map constructed from bottom-up and top-down information bottom-up activations FIGURE 2. Sample display from Experiment 1 of Maljkovic and Nakayama (1994) response or activation indicating the presence of a particular feature. Most models assume that responses are stronger at locations with high local feature contrast, consistent with neurophysiological data, e.g., the response of a red feature detector to a red object is stronger if the object is surrounded by green objects. The saliency map is obtained by taking a sum of bottom-up activations from the feature maps. The bottom-up activations are modulated by a top-down gain that specifies the contribution of a particular map to saliency in the current task and environment. Wolfe (1994) describes a heuristic algorithm for determining appropriate gains in a visual search task, where the goal is to detect a target object among distractor objects. Wolfe proposes that maps encoding features that discriminate between target and distractors have higher gains, and to be consistent with the data, he proposes limits on the magnitude of gain modulation and the number of gains that can be modulated. More recently, Wolfe et al. (2003) have been explicit in proposing optimization as a principle for setting gains given the task definition and stimulus environment. One aspect of optimizing attentional control involves configuring the attentional system to perform a given task; for example, in a visual search task for a red vertical target among green vertical and red horizontal distractors, the task definition should result in a higher gain for red and vertical feature maps than for other feature maps. However, there is a more subtle form of gain modulation, which depends on the statistics of display environments. For example, if green vertical distractors predominate, then red is a better discriminative cue than vertical; and if red horizontal distractors predominate, then vertical is a better discriminative cue than red. In this paper, we propose a model that encodes statistics of the environment in order to allow for optimization of attentional control to the structure of the environment. Our model is designed to address a key set of behavioral data, which we describe next. 1.1 Attentional priming phenomena Psychological studies involve a sequence of experimental trials that begin with a stimulus presentation and end with a response from the human participant. Typically, trial order is randomized, and the context preceding a trial is ignored. However, in sequential studies, performance is examined on one trial contingent on the past history of trials. These sequential studies explore how experience influences future performance. Consider a the sequential attentional task of Maljkovic and Nakayama (1994). On each trial, the stimulus display (Figure 2) consists of three notched diamonds, one a singleton in color—either green among red or red among green. The task is to report whether the singleton diamond, referred to as the target, is notched on the left or the right. The task is easy because the singleton pops out, i.e., the time to locate the singleton does not depend on the number of diamonds in the display. Nonetheless, the response time significantly depends on the sequence of trials leading up to the current trial: If the target is the same color on the cur- rent trial as on the previous trial, response time is roughly 100 ms faster than if the target is a different color on the current trial. Considering that response times are on the order of 700 ms, this effect, which we term attentional priming, is gigantic in the scheme of psychological phenomena. 2 ATTENTIONAL CONTROL AS ADAPTATION TO THE STATISTICS OF THE ENVIRONMENT We interpret the phenomenon of attentional priming via a particular perspective on attentional control, which can be summarized in two bullets. • The perceptual system dynamically constructs a probabilistic model of the environment based on its past experience. • Control parameters of the attentional system are tuned so as to optimize performance under the current environmental model. The primary focus of this paper is the environmental model, but we first discuss the nature of performance optimization. The role of attention is to make processing of some stimuli more efficient, and consequently, the processing of other stimuli less efficient. For example, if the gain on the red feature map is turned up, processing will be efficient for red items, but competition from red items will reduce the efficiency for green items. Thus, optimal control should tune the system for the most likely states of the world by minimizing an objective function such as: J(g) = ∑ P ( e )RT g ( e ) (1) e where g is a vector of top-down gains, e is an index over environmental states, P(.) is the probability of an environmental state, and RTg(.) is the expected response time—assuming a constant error rate—to the environmental state under gains g. Determining the optimal gains is a challenge because every gain setting will result in facilitation of responses to some environmental states but hindrance of responses to other states. The optimal control problem could be solved via direct reinforcement learning, but the rapidity of human learning makes this possibility unlikely: In a variety of experimental tasks, evidence suggests that adaptation to a new task or environment can occur in just one or two trials (e.g., Rogers & Monsell, 1996). Model-based reinforcement learning is an attractive alternative, because given a model, optimization can occur without further experience in the real world. Although the number of real-world trials necessary to achieve a given level of performance is comparable for direct and model-based reinforcement learning in stationary environments (Kearns & Singh, 1999), naturalistic environments can be viewed as highly nonstationary. In such a situation, the framework we suggest is well motivated: After each experience, the environment model is updated. The updated environmental model is then used to retune the attentional system. In this paper, we propose a particular model of the environment suitable for visual search tasks. Rather than explicitly modeling the optimization of attentional control by setting gains, we assume that the optimization process will serve to minimize Equation 1. Because any gain adjustment will facilitate performance in some environmental states and hinder performance in others, an optimized control system should obtain faster reaction times for more probable environmental states. This assumption allows us to explain experimental results in a minimal, parsimonious framework. 3 MODELING THE ENVIRONMENT Focusing on the domain of visual search, we characterize the environment in terms of a probability distribution over configurations of target and distractor features. We distinguish three classes of features: defining, reported, and irrelevant. To explain these terms, consider the task of searching a display of size varying, colored, notched diamonds (Figure 2), with the task of detecting the singleton in color and judging the notch location. Color is the defining feature, notch location is the reported feature, and size is an irrelevant feature. To simplify the exposition, we treat all features as having discrete values, an assumption which is true of the experimental tasks we model. We begin by considering displays containing a single target and a single distractor, and shortly generalize to multidistractor displays. We use the framework of Bayesian networks to characterize the environment. Each feature of the target and distractor is a discrete random variable, e.g., Tcolor for target color and Dnotch for the location of the notch on the distractor. The Bayes net encodes the probability distribution over environmental states; in our working example, this distribution is P(Tcolor, Tsize, Tnotch, Dcolor, Dsize, Dnotch). The structure of the Bayes net specifies the relationships among the features. The simplest model one could consider would be to treat the features as independent, illustrated in Figure 3a for singleton-color search task. The opposite extreme would be the full joint distribution, which could be represented by a look up table indexed by the six features, or by the cascading Bayes net architecture in Figure 3b. The architecture we propose, which we’ll refer to as the dominance model (Figure 3c), has an intermediate dependency structure, and expresses the joint distribution as: P(Tcolor)P(Dcolor |Tcolor)P(Tsize |Tcolor)P(Tnotch |Tcolor)P(Dsize |Dcolor)P(Dnotch |Tcolor). The structured model is constructed based on three rules. 1. The defining feature of the target is at the root of the tree. 2. The defining feature of the distractor is conditionally dependent on the defining feature of the target. We refer to this rule as dominance of the target over the distractor. 3. The reported and irrelevant features of target (distractor) are conditionally dependent on the defining feature of the target (distractor). We refer to this rule as dominance of the defining feature over nondefining features. As we will demonstrate, the dominance model produces a parsimonious account of a wide range of experimental data. 3.1 Updating the environment model The model’s parameters are the conditional distributions embodied in the links. In the example of Figure 3c with binary random variables, the model has 11 parameters. However, these parameters are determined by the environment: To be adaptive in nonstationary environments, the model must be updated following each experienced state. We propose a simple exponentially weighted averaging approach. For two variables V and W with observed values v and w on trial t, a conditional distribution, P t ( V = u W = w ) = δ uv , is (a) Tcolor Dcolor Tsize Tnotch (b) Tcolor Dcolor Dsize Tsize Dnotch Tnotch (c) Tcolor Dcolor Dsize Tsize Dsize Dnotch Tnotch Dnotch FIGURE 3. Three models of a visual-search environment with colored, notched, size-varying diamonds. (a) feature-independence model; (b) full-joint model; (c) dominance model. defined, where δ is the Kronecker delta. The distribution representing the environment E following trial t, denoted P t , is then updated as follows: E E P t ( V = u W = w ) = αP t – 1 ( V = u W = w ) + ( 1 – α )P t ( V = u W = w ) (2) for all u, where α is a memory constant. Note that no update is performed for values of W other than w. An analogous update is performed for unconditional distributions. E How the model is initialized—i.e., specifying P 0 —is irrelevant, because all experimental tasks that we model, participants begin the experiment with many dozens of practice trials. E Data is not collected during practice trials. Consequently, any transient effects of P 0 do E not impact the results. In our simulations, we begin with a uniform distribution for P 0 , and include practice trials as in the human studies. Thus far, we’ve assumed a single target and a single distractor. The experiments that we model involve multiple distractors. The simple extension we require to handle multiple distractors is to define a frequentist probability for each distractor feature V, P t ( V = v W = w ) = C vw ⁄ C w , where C vw is the count of co-occurrences of feature values v and w among the distractors, and C w is the count of w. Our model is extremely simple. Given a description of the visual search task and environment, the model has only a single degree of freedom, α . In all simulations, we fix α = 0.75 ; however, the choice of α does not qualitatively impact any result. 4 SIMULATIONS In this section, we show that the model can explain a range of data from four different experiments examining attentional priming. All experiments measure response times of participants. On each trial, the model can be used to obtain a probability of the display configuration (the environmental state) on that trial, given the history of trials to that point. Our critical assumption—as motivated earlier—is that response times monotonically decrease with increasing probability, indicating that visual information processing is better configured for more likely environmental states. The particular relationship we assume is that response times are linear in log probability. This assumption yields long response time tails, as are observed in all human studies. 4.1 Maljkovic and Nakayama (1994, Experiment 5) In this experiment, participants were asked to search for a singleton in color in a display of three red or green diamonds. Each diamond was notched on either the left or right side, and the task was to report the side of the notch on the color singleton. The well-practiced participants made very few errors. Reaction time (RT) was examined as a function of whether the target on a given trial is the same or different color as the target on trial n steps back or ahead. Figure 4 shows the results, with the human RTs in the left panel and the simulation log probabilities in the right panel. The horizontal axis represents n. Both graphs show the same outcome: repetition of target color facilitates performance. This influence lasts only for a half dozen trials, with an exponentially decreasing influence further into the past. In the model, this decreasing influence is due to the exponential decay of recent history (Equation 2). Figure 4 also shows that—as expected—the future has no influence on the current trial. 4.2 Maljkovic and Nakayama (1994, Experiment 8) In the previous experiment, it is impossible to determine whether facilitation is due to repetition of the target’s color or the distractor’s color, because the display contains only two colors, and therefore repetition of target color implies repetition of distractor color. To unconfound these two potential factors, an experiment like the previous one was con- ducted using four distinct colors, allowing one to examine the effect of repeating the target color while varying the distractor color, and vice versa. The sequence of trials was composed of subsequences of up-to-six consecutive trials with either the target or distractor color held constant while the other color was varied trial to trial. Following each subsequence, both target and distractors were changed. Figure 5 shows that for both humans and the simulation, performance improves toward an asymptote as the number of target and distractor repetitions increases; in the model, the asymptote is due to the probability of the repeated color in the environment model approaching 1.0. The performance improvement is greater for target than distractor repetition; in the model, this difference is due to the dominance of the defining feature of the target over the defining feature of the distractor. 4.3 Huang, Holcombe, and Pashler (2004, Experiment 1) Huang et al. (2004) and Hillstrom (2000) conducted studies to determine whether repetitions of one feature facilitate performance independently of repetitions of another feature. In the Huang et al. study, participants searched for a singleton in size in a display consisting of lines that were short and long, slanted left or right, and colored white or black. The reported feature was target slant. Slant, size, and color were uncorrelated. Huang et al. discovered that repeating an irrelevant feature (color or orientation) facilitated performance, but only when the defining feature (size) was repeated. As shown in Figure 6, the model replicates human performance, due to the dominance of the defining feature over the reported and irrelevant features. 4.4 Wolfe, Butcher, Lee, and Hyde (2003, Experiment 1) In an empirical tour-de-force, Wolfe et al. (2003) explored singleton search over a range of environments. The task is to detect the presence or absence of a singleton in displays conHuman data Different Color 600 Different Color 590 580 570 15 13 11 9 7 Past 5 3.2 3 Same Color 2.8 Same Color 560 550 Simulation 3.4 log(P(trial)) Reaction Time (msec) 610 3 1 +1 +3 +5 Future 2.6 +7 15 13 Relative Trial Number 11 9 7 Past 5 3 1 +1 +3 +5 Future +7 Relative Trial Number FIGURE 4. Experiment 5 of Maljkovic and Nakayama (1994): performance on a given trial conditional on the color of the target on a previous or subsequent trial. Human data is from subject KN. 650 6 Distractors Same 630 5.5 log(P(trial)) FIGURE 5. Experiment 8 of Maljkovic and Nakayama (1994). (left panel) human data, average of subjects KN and SS; (right panel) simulation Reaction Time (msec) 640 620 Target Same 610 5 Distractors Same 4.5 4 600 Target Same 3.5 590 3 580 1 2 3 4 5 1 6 4 5 6 4 1000 Size Alternate Size Alternate log(P(trial)) Reaction Time (msec) 3 4.2 1050 FIGURE 6. Experiment 1 of Huang, Holcombe, & Pashler (2004). (left panel) human data; (right panel) simulation 2 Order in Sequence Order in Sequence 950 3.8 3.6 3.4 900 3.2 Size Repeat 850 Size Repeat 3 Color Repeat Color Alternate Color Repeat Color Alternate sisting of colored (red or green), oriented (horizontal or vertical) lines. Target-absent trials were used primarily to ensure participants were searching the display. The experiment examined seven experimental conditions, which varied in the amount of uncertainty as to the target identity. The essential conditions, from least to most uncertainty, are: blocked (e.g., target always red vertical among green horizontals), mixed feature (e.g., target always a color singleton), mixed dimension (e.g., target either red or vertical), and fully mixed (target could be red, green, vertical, or horizontal). With this design, one can ascertain how uncertainty in the environment and in the target definition influence task difficulty. Because the defining feature in this experiment could be either color or orientation, we modeled the environment with two Bayes nets—one color dominant and one orientation dominant—and performed model averaging. A comparison of Figures 7a and 7b show a correspondence between human RTs and model predictions. Less uncertainty in the environment leads to more efficient performance. One interesting result from the model is its prediction that the mixed-feature condition is easier than the fully-mixed condition; that is, search is more efficient when the dimension (i.e., color vs. orientation) of the singleton is known, even though the model has no abstract representation of feature dimensions, only feature values. 4.5 Optimal adaptation constant In all simulations so far, we fixed the memory constant. From the human data, it is clear that memory for recent experience is relatively short lived, on the order of a half dozen trials (e.g., left panel of Figure 4). In this section we provide a rational argument for the short duration of memory in attentional control. Figure 7c shows mean negative log probability in each condition of the Wolfe et al. (2003) experiment, as a function of α . To assess these probabilities, for each experimental condition, the model was initialized so that all of the conditional distributions were uniform, and then a block of trials was run. Log probability for all trials in the block was averaged. The negative log probability (y axis of the Figure) is a measure of the model’s misprediction of the next trial in the sequence. For complex environments, such as the fully-mixed condition, a small memory constant is detrimental: With rapid memory decay, the effective history of trials is a high-variance sample of the distribution of environmental states. For simple environments, a large memory constant is detrimental: With slow memory decay, the model does not transition quickly from the initial environmental model to one that reflects the statistics of a new environment. Thus, the memory constant is constrained by being large enough that the environment model can hold on to sufficient history to represent complex environments, and by being small enough that the model adapts quickly to novel environments. If the conditions in Wolfe et al. give some indication of the range of naturalistic environments an agent encounters, we have a rational account of why attentional priming is so short lived. Whether priming lasts 2 trials or 20, the surprising empirical result is that it does not last 200 or 2000 trials. Our rational argument provides a rough insight into this finding. (a) fully mixed mixed feature mixed dimension blocked 460 (c) Simulation fully mixed mixed feature mixed dimension blocked 4 5 420 log(P(trial)) 440 2 Blocked Red or Vertical Blocked Red and Vertical Mixed Feature Mixed Dimension Fully Mixed 4 3 log(P(trial)) reaction time (msec) (b) Human Data 480 3 2 1 400 1 380 0 360 0 red or vert red and vert target type red or vert red and vert target type 0 0.5 0.8 0.9 0.95 0.98 Memory Constant FIGURE 7. (a) Human data for Wolfe et al. (2003), Experiment 1; (b) simulation; (c) misprediction of model (i.e., lower y value = better) as a function of α for five experimental condition 5 DISCUSSION The psychological literature contains two opposing accounts of attentional priming and its relation to attentional control. Huang et al. (2004) and Hillstrom (2000) propose an episodic account in which a distinct memory trace—representing the complete configuration of features in the display—is laid down for each trial, and priming depends on configural similarity of the current trial to previous trials. Alternatively, Maljkovic and Nakayama (1994) and Wolfe et al. (2003) propose a feature-strengthening account in which detection of a feature on one trial increases its ability to attract attention on subsequent trials, and priming is proportional to the number of overlapping features from one trial to the next. The episodic account corresponds roughly to the full joint model (Figure 3b), and the feature-strengthening account corresponds roughly to the independence model (Figure 3a). Neither account is adequate to explain the range of data we presented. However, an intermediate account, the dominance model (Figure 3c), is not only sufficient, but it offers a parsimonious, rational explanation. Beyond the model’s basic assumptions, it has only one free parameter, and can explain results from diverse experimental paradigms. The model makes a further theoretical contribution. Wolfe et al. distinguish the environments in their experiment in terms of the amount of top-down control available, implying that different mechanisms might be operating in different environments. However, in our account, top-down control is not some substance distributed in different amounts depending on the nature of the environment. Our account treats all environments uniformly, relying on attentional control to adapt to the environment at hand. We conclude with two limitations of the present work. First, our account presumes a particular network architecture, instead of a more elegant Bayesian approach that specifies priors over architectures, and performs automatic model selection via the sequence of trials. We did explore such a Bayesian approach, but it was unable to explain the data. Second, at least one finding in the literature is problematic for the model. Hillstrom (2000) occasionally finds that RTs slow when an irrelevant target feature is repeated but the defining target feature is not. However, because this effect is observed only in some experiments, it is likely that any model would require elaboration to explain the variability. ACKNOWLEDGEMENTS We thank Jeremy Wolfe for providing the raw data from his experiment for reanalysis. This research was funded by NSF BCS Award 0339103. REFERENCES Huang, L, Holcombe, A. O., & Pashler, H. (2004). Repetition priming in visual search: Episodic retrieval, not feature priming. Memory & Cognition, 32, 12–20. Hillstrom, A. P. (2000). Repetition effects in visual search. Perception & Psychophysics, 62, 800-817. Itti, L., Koch, C., & Niebur, E. (1998). A model of saliency-based visual attention for rapid scene analysis. IEEE Trans. Pattern Analysis & Machine Intelligence, 20, 1254–1259. Kearns, M., & Singh, S. (1999). Finite-sample convergence rates for Q-learning and indirect algorithms. In Advances in Neural Information Processing Systems 11 (pp. 996–1002). Cambridge, MA: MIT Press. Koch, C. and Ullman, S. (1985). Shifts in selective visual attention: towards the underlying neural circuitry. Human Neurobiology, 4, 219–227. Maljkovic, V., & Nakayama, K. (1994). Priming of pop-out: I. Role of features. Mem. & Cognition, 22, 657-672. Mozer, M. C. (1991). The perception of multiple objects: A connectionist approach. Cambridge, MA: MIT Press. Rogers, R. D., & Monsell, S. (1995). The cost of a predictable switch between simple cognitive tasks. Journal of Experimental Psychology: General, 124, 207–231. Wolfe, J.M. (1994). Guided Search 2.0: A Revised Model of Visual Search. Psych. Bull. & Rev., 1, 202–238. Wolfe, J. S., Butcher, S. J., Lee, C., & Hyde, M. (2003). Changing your mind: on the contributions of top-down and bottom-up guidance in visual search for feature singletons. Journal of Exptl. Psychology: Human Perception & Performance, 29, 483-502.
Reference: text
sentIndex sentText sentNum sentScore
1 of Psychology University of Iowa Iowa City, IA 52242 USA Abstract Theories of visual attention commonly posit that early parallel processes extract conspicuous features such as color contrast and motion from the visual field. [sent-9, score-0.6]
2 These features are then combined into a saliency map, and attention is directed to the most salient regions first. [sent-10, score-0.244]
3 Top-down attentional control is achieved by modulating the contribution of different feature types to the saliency map. [sent-11, score-0.584]
4 A key source of data concerning attentional control comes from behavioral studies in which the effect of recent experience is examined as individuals repeatedly perform a perceptual discrimination task (e. [sent-12, score-0.496]
5 The robust finding is that repetition of features of recent trials (e. [sent-16, score-0.278]
6 Under the assumption that attentional control operates so as to make performance more efficient for more likely environmental states, we obtain parsimonious explanations for data from four different experiments. [sent-21, score-0.654]
7 Further, our model provides a rational explanation for why the influence of past experience on attentional control is short lived. [sent-22, score-0.606]
8 Most theories of attention posit that some visual information processing is performed preattentively and in parallel across the visual field. [sent-30, score-0.284]
9 This processing extracts primitive visual features such as color and motion, which provide the bottom-up cues for attentional guidance. [sent-31, score-0.741]
10 Most psychologically and neurobiologically motivated models propose a very similar architecture in which information from bottom-up and top-down sources combines in a saliency (or activation) map (e. [sent-35, score-0.189]
11 The saliency map indicates, for each location in the visual field, the relative importance of that location. [sent-39, score-0.244]
12 The visual image is analyzed to extract maps of primitive features such as color and orientation. [sent-42, score-0.452]
13 Associated with each location in a map is a scalar visual image horizontal primitive feature maps vertical green top-down gains red saliency map FIGURE 1. [sent-43, score-0.885]
14 An attentional saliency map constructed from bottom-up and top-down information bottom-up activations FIGURE 2. [sent-44, score-0.479]
15 , the response of a red feature detector to a red object is stronger if the object is surrounded by green objects. [sent-48, score-0.539]
16 The saliency map is obtained by taking a sum of bottom-up activations from the feature maps. [sent-49, score-0.303]
17 The bottom-up activations are modulated by a top-down gain that specifies the contribution of a particular map to saliency in the current task and environment. [sent-50, score-0.328]
18 Wolfe (1994) describes a heuristic algorithm for determining appropriate gains in a visual search task, where the goal is to detect a target object among distractor objects. [sent-51, score-0.59]
19 Wolfe proposes that maps encoding features that discriminate between target and distractors have higher gains, and to be consistent with the data, he proposes limits on the magnitude of gain modulation and the number of gains that can be modulated. [sent-52, score-0.453]
20 For example, if green vertical distractors predominate, then red is a better discriminative cue than vertical; and if red horizontal distractors predominate, then vertical is a better discriminative cue than red. [sent-57, score-0.823]
21 In this paper, we propose a model that encodes statistics of the environment in order to allow for optimization of attentional control to the structure of the environment. [sent-58, score-0.516]
22 1 Attentional priming phenomena Psychological studies involve a sequence of experimental trials that begin with a stimulus presentation and end with a response from the human participant. [sent-61, score-0.388]
23 Typically, trial order is randomized, and the context preceding a trial is ignored. [sent-62, score-0.438]
24 However, in sequential studies, performance is examined on one trial contingent on the past history of trials. [sent-63, score-0.332]
25 Consider a the sequential attentional task of Maljkovic and Nakayama (1994). [sent-65, score-0.337]
26 On each trial, the stimulus display (Figure 2) consists of three notched diamonds, one a singleton in color—either green among red or red among green. [sent-66, score-0.702]
27 The task is to report whether the singleton diamond, referred to as the target, is notched on the left or the right. [sent-67, score-0.297]
28 The task is easy because the singleton pops out, i. [sent-68, score-0.202]
29 , the time to locate the singleton does not depend on the number of diamonds in the display. [sent-70, score-0.199]
30 Nonetheless, the response time significantly depends on the sequence of trials leading up to the current trial: If the target is the same color on the cur- rent trial as on the previous trial, response time is roughly 100 ms faster than if the target is a different color on the current trial. [sent-71, score-1.315]
31 Considering that response times are on the order of 700 ms, this effect, which we term attentional priming, is gigantic in the scheme of psychological phenomena. [sent-72, score-0.342]
32 2 ATTENTIONAL CONTROL AS ADAPTATION TO THE STATISTICS OF THE ENVIRONMENT We interpret the phenomenon of attentional priming via a particular perspective on attentional control, which can be summarized in two bullets. [sent-73, score-0.743]
33 • Control parameters of the attentional system are tuned so as to optimize performance under the current environmental model. [sent-75, score-0.445]
34 For example, if the gain on the red feature map is turned up, processing will be efficient for red items, but competition from red items will reduce the efficiency for green items. [sent-78, score-0.8]
35 Thus, optimal control should tune the system for the most likely states of the world by minimizing an objective function such as: J(g) = ∑ P ( e )RT g ( e ) (1) e where g is a vector of top-down gains, e is an index over environmental states, P(. [sent-79, score-0.23]
36 ) is the expected response time—assuming a constant error rate—to the environmental state under gains g. [sent-81, score-0.297]
37 Determining the optimal gains is a challenge because every gain setting will result in facilitation of responses to some environmental states but hindrance of responses to other states. [sent-82, score-0.325]
38 The optimal control problem could be solved via direct reinforcement learning, but the rapidity of human learning makes this possibility unlikely: In a variety of experimental tasks, evidence suggests that adaptation to a new task or environment can occur in just one or two trials (e. [sent-83, score-0.445]
39 Although the number of real-world trials necessary to achieve a given level of performance is comparable for direct and model-based reinforcement learning in stationary environments (Kearns & Singh, 1999), naturalistic environments can be viewed as highly nonstationary. [sent-87, score-0.316]
40 The updated environmental model is then used to retune the attentional system. [sent-89, score-0.445]
41 In this paper, we propose a particular model of the environment suitable for visual search tasks. [sent-90, score-0.298]
42 Rather than explicitly modeling the optimization of attentional control by setting gains, we assume that the optimization process will serve to minimize Equation 1. [sent-91, score-0.363]
43 Because any gain adjustment will facilitate performance in some environmental states and hinder performance in others, an optimized control system should obtain faster reaction times for more probable environmental states. [sent-92, score-0.499]
44 3 MODELING THE ENVIRONMENT Focusing on the domain of visual search, we characterize the environment in terms of a probability distribution over configurations of target and distractor features. [sent-94, score-0.606]
45 To explain these terms, consider the task of searching a display of size varying, colored, notched diamonds (Figure 2), with the task of detecting the singleton in color and judging the notch location. [sent-96, score-0.865]
46 Color is the defining feature, notch location is the reported feature, and size is an irrelevant feature. [sent-97, score-0.32]
47 Each feature of the target and distractor is a discrete random variable, e. [sent-101, score-0.47]
48 , Tcolor for target color and Dnotch for the location of the notch on the distractor. [sent-103, score-0.52]
49 The defining feature of the target is at the root of the tree. [sent-111, score-0.481]
50 The defining feature of the distractor is conditionally dependent on the defining feature of the target. [sent-113, score-0.803]
51 We refer to this rule as dominance of the target over the distractor. [sent-114, score-0.292]
52 The reported and irrelevant features of target (distractor) are conditionally dependent on the defining feature of the target (distractor). [sent-116, score-0.745]
53 We refer to this rule as dominance of the defining feature over nondefining features. [sent-117, score-0.429]
54 As we will demonstrate, the dominance model produces a parsimonious account of a wide range of experimental data. [sent-118, score-0.215]
55 For two variables V and W with observed values v and w on trial t, a conditional distribution, P t ( V = u W = w ) = δ uv , is (a) Tcolor Dcolor Tsize Tnotch (b) Tcolor Dcolor Dsize Tsize Dnotch Tnotch (c) Tcolor Dcolor Dsize Tsize Dsize Dnotch Tnotch Dnotch FIGURE 3. [sent-124, score-0.219]
56 The distribution representing the environment E following trial t, denoted P t , is then updated as follows: E E P t ( V = u W = w ) = αP t – 1 ( V = u W = w ) + ( 1 – α )P t ( V = u W = w ) (2) for all u, where α is a memory constant. [sent-128, score-0.448]
57 The simple extension we require to handle multiple distractors is to define a frequentist probability for each distractor feature V, P t ( V = v W = w ) = C vw ⁄ C w , where C vw is the count of co-occurrences of feature values v and w among the distractors, and C w is the count of w. [sent-139, score-0.519]
58 Given a description of the visual search task and environment, the model has only a single degree of freedom, α . [sent-141, score-0.193]
59 4 SIMULATIONS In this section, we show that the model can explain a range of data from four different experiments examining attentional priming. [sent-144, score-0.336]
60 On each trial, the model can be used to obtain a probability of the display configuration (the environmental state) on that trial, given the history of trials to that point. [sent-146, score-0.382]
61 Our critical assumption—as motivated earlier—is that response times monotonically decrease with increasing probability, indicating that visual information processing is better configured for more likely environmental states. [sent-147, score-0.305]
62 1 Maljkovic and Nakayama (1994, Experiment 5) In this experiment, participants were asked to search for a singleton in color in a display of three red or green diamonds. [sent-151, score-0.833]
63 Each diamond was notched on either the left or right side, and the task was to report the side of the notch on the color singleton. [sent-152, score-0.524]
64 Reaction time (RT) was examined as a function of whether the target on a given trial is the same or different color as the target on trial n steps back or ahead. [sent-154, score-1.089]
65 Both graphs show the same outcome: repetition of target color facilitates performance. [sent-157, score-0.576]
66 This influence lasts only for a half dozen trials, with an exponentially decreasing influence further into the past. [sent-158, score-0.251]
67 To unconfound these two potential factors, an experiment like the previous one was con- ducted using four distinct colors, allowing one to examine the effect of repeating the target color while varying the distractor color, and vice versa. [sent-163, score-0.706]
68 The sequence of trials was composed of subsequences of up-to-six consecutive trials with either the target or distractor color held constant while the other color was varied trial to trial. [sent-164, score-1.324]
69 Following each subsequence, both target and distractors were changed. [sent-165, score-0.28]
70 Figure 5 shows that for both humans and the simulation, performance improves toward an asymptote as the number of target and distractor repetitions increases; in the model, the asymptote is due to the probability of the repeated color in the environment model approaching 1. [sent-166, score-0.894]
71 The performance improvement is greater for target than distractor repetition; in the model, this difference is due to the dominance of the defining feature of the target over the defining feature of the distractor. [sent-168, score-1.267]
72 study, participants searched for a singleton in size in a display consisting of lines that were short and long, slanted left or right, and colored white or black. [sent-173, score-0.336]
73 discovered that repeating an irrelevant feature (color or orientation) facilitated performance, but only when the defining feature (size) was repeated. [sent-177, score-0.47]
74 As shown in Figure 6, the model replicates human performance, due to the dominance of the defining feature over the reported and irrelevant features. [sent-178, score-0.545]
75 (2003) explored singleton search over a range of environments. [sent-181, score-0.203]
76 The task is to detect the presence or absence of a singleton in displays conHuman data Different Color 600 Different Color 590 580 570 15 13 11 9 7 Past 5 3. [sent-182, score-0.202]
77 Experiment 5 of Maljkovic and Nakayama (1994): performance on a given trial conditional on the color of the target on a previous or subsequent trial. [sent-187, score-0.663]
78 The experiment examined seven experimental conditions, which varied in the amount of uncertainty as to the target identity. [sent-203, score-0.284]
79 , target always red vertical among green horizontals), mixed feature (e. [sent-206, score-0.712]
80 , target either red or vertical), and fully mixed (target could be red, green, vertical, or horizontal). [sent-210, score-0.432]
81 With this design, one can ascertain how uncertainty in the environment and in the target definition influence task difficulty. [sent-211, score-0.508]
82 Because the defining feature in this experiment could be either color or orientation, we modeled the environment with two Bayes nets—one color dominant and one orientation dominant—and performed model averaging. [sent-212, score-1.083]
83 Less uncertainty in the environment leads to more efficient performance. [sent-214, score-0.235]
84 orientation) of the singleton is known, even though the model has no abstract representation of feature dimensions, only feature values. [sent-218, score-0.38]
85 From the human data, it is clear that memory for recent experience is relatively short lived, on the order of a half dozen trials (e. [sent-221, score-0.329]
86 In this section we provide a rational argument for the short duration of memory in attentional control. [sent-224, score-0.434]
87 The negative log probability (y axis of the Figure) is a measure of the model’s misprediction of the next trial in the sequence. [sent-229, score-0.257]
88 For complex environments, such as the fully-mixed condition, a small memory constant is detrimental: With rapid memory decay, the effective history of trials is a high-variance sample of the distribution of environmental states. [sent-230, score-0.454]
89 For simple environments, a large memory constant is detrimental: With slow memory decay, the model does not transition quickly from the initial environmental model to one that reflects the statistics of a new environment. [sent-231, score-0.308]
90 Thus, the memory constant is constrained by being large enough that the environment model can hold on to sufficient history to represent complex environments, and by being small enough that the model adapts quickly to novel environments. [sent-232, score-0.273]
91 give some indication of the range of naturalistic environments an agent encounters, we have a rational account of why attentional priming is so short lived. [sent-234, score-0.691]
92 Whether priming lasts 2 trials or 20, the surprising empirical result is that it does not last 200 or 2000 trials. [sent-235, score-0.305]
93 , lower y value = better) as a function of α for five experimental condition 5 DISCUSSION The psychological literature contains two opposing accounts of attentional priming and its relation to attentional control. [sent-246, score-0.743]
94 (2004) and Hillstrom (2000) propose an episodic account in which a distinct memory trace—representing the complete configuration of features in the display—is laid down for each trial, and priming depends on configural similarity of the current trial to previous trials. [sent-248, score-0.591]
95 (2003) propose a feature-strengthening account in which detection of a feature on one trial increases its ability to attract attention on subsequent trials, and priming is proportional to the number of overlapping features from one trial to the next. [sent-250, score-0.894]
96 distinguish the environments in their experiment in terms of the amount of top-down control available, implying that different mechanisms might be operating in different environments. [sent-257, score-0.239]
97 Our account treats all environments uniformly, relying on attentional control to adapt to the environment at hand. [sent-259, score-0.646]
98 Hillstrom (2000) occasionally finds that RTs slow when an irrelevant target feature is repeated but the defining target feature is not. [sent-264, score-0.814]
99 Repetition priming in visual search: Episodic retrieval, not feature priming. [sent-272, score-0.374]
100 Changing your mind: on the contributions of top-down and bottom-up guidance in visual search for feature singletons. [sent-332, score-0.295]
wordName wordTfidf (topN-words)
[('attentional', 0.289), ('color', 0.272), ('trial', 0.219), ('defining', 0.196), ('tcolor', 0.19), ('distractor', 0.185), ('wolfe', 0.176), ('target', 0.172), ('priming', 0.165), ('environmental', 0.156), ('singleton', 0.154), ('environment', 0.153), ('nakayama', 0.152), ('red', 0.151), ('repetition', 0.132), ('maljkovic', 0.132), ('dominance', 0.12), ('dcolor', 0.114), ('dnotch', 0.114), ('feature', 0.113), ('mixed', 0.109), ('saliency', 0.108), ('distractors', 0.108), ('trials', 0.102), ('visual', 0.096), ('vertical', 0.096), ('dsize', 0.095), ('notched', 0.095), ('tnotch', 0.095), ('tsize', 0.095), ('attention', 0.092), ('influence', 0.09), ('gains', 0.088), ('environments', 0.088), ('huang', 0.084), ('efficient', 0.082), ('display', 0.08), ('experiment', 0.077), ('hillstrom', 0.076), ('notch', 0.076), ('memory', 0.076), ('control', 0.074), ('reaction', 0.072), ('green', 0.071), ('rational', 0.069), ('human', 0.068), ('blocked', 0.066), ('holcombe', 0.057), ('msec', 0.056), ('participants', 0.056), ('response', 0.053), ('parsimonious', 0.053), ('vert', 0.05), ('experience', 0.05), ('pashler', 0.049), ('rts', 0.049), ('specifies', 0.049), ('search', 0.049), ('task', 0.048), ('irrelevant', 0.048), ('explain', 0.047), ('colored', 0.046), ('diamonds', 0.045), ('episodic', 0.045), ('definition', 0.045), ('history', 0.044), ('features', 0.044), ('panel', 0.043), ('account', 0.042), ('activations', 0.042), ('horizontal', 0.042), ('architecture', 0.041), ('gain', 0.041), ('map', 0.04), ('alternate', 0.04), ('simulation', 0.04), ('facilitation', 0.04), ('primitive', 0.04), ('et', 0.039), ('asymptote', 0.038), ('boulder', 0.038), ('detrimental', 0.038), ('hyde', 0.038), ('lasts', 0.038), ('misprediction', 0.038), ('monsell', 0.038), ('naturalistic', 0.038), ('predominate', 0.038), ('mozer', 0.038), ('contributions', 0.037), ('repetitions', 0.036), ('koch', 0.036), ('examined', 0.035), ('repeat', 0.035), ('past', 0.034), ('colorado', 0.033), ('diamond', 0.033), ('iowa', 0.033), ('dozen', 0.033), ('butcher', 0.033)]
simIndex simValue paperId paperTitle
same-paper 1 1.0000007 194 nips-2005-Top-Down Control of Visual Attention: A Rational Account
Author: Michael Shettel, Shaun Vecera, Michael C. Mozer
Abstract: Theories of visual attention commonly posit that early parallel processes extract conspicuous features such as color contrast and motion from the visual field. These features are then combined into a saliency map, and attention is directed to the most salient regions first. Top-down attentional control is achieved by modulating the contribution of different feature types to the saliency map. A key source of data concerning attentional control comes from behavioral studies in which the effect of recent experience is examined as individuals repeatedly perform a perceptual discrimination task (e.g., “what shape is the odd-colored object?”). The robust finding is that repetition of features of recent trials (e.g., target color) facilitates performance. We view this facilitation as an adaptation to the statistical structure of the environment. We propose a probabilistic model of the environment that is updated after each trial. Under the assumption that attentional control operates so as to make performance more efficient for more likely environmental states, we obtain parsimonious explanations for data from four different experiments. Further, our model provides a rational explanation for why the influence of past experience on attentional control is short lived. 1 INTRODUCTION The brain does not have the computational capacity to fully process the massive quantity of information provided by the eyes. Selective attention operates to filter the spatiotemporal stream to a manageable quantity. Key to understanding the nature of attention is discovering the algorithm governing selection, i.e., understanding what information will be selected and what will be suppressed. Selection is influenced by attributes of the spatiotemporal stream, often referred to as bottom-up contributions to attention. For example, attention is drawn to abrupt onsets, motion, and regions of high contrast in brightness and color. Most theories of attention posit that some visual information processing is performed preattentively and in parallel across the visual field. This processing extracts primitive visual features such as color and motion, which provide the bottom-up cues for attentional guidance. However, attention is not driven willy nilly by these cues. The deployment of attention can be modulated by task instructions, current goals, and domain knowledge, collectively referred to as top-down contributions to attention. How do bottom-up and top-down contributions to attention interact? Most psychologically and neurobiologically motivated models propose a very similar architecture in which information from bottom-up and top-down sources combines in a saliency (or activation) map (e.g., Itti et al., 1998; Koch & Ullman, 1985; Mozer, 1991; Wolfe, 1994). The saliency map indicates, for each location in the visual field, the relative importance of that location. Attention is drawn to the most salient locations first. Figure 1 sketches the basic architecture that incorporates bottom-up and top-down contributions to the saliency map. The visual image is analyzed to extract maps of primitive features such as color and orientation. Associated with each location in a map is a scalar visual image horizontal primitive feature maps vertical green top-down gains red saliency map FIGURE 1. An attentional saliency map constructed from bottom-up and top-down information bottom-up activations FIGURE 2. Sample display from Experiment 1 of Maljkovic and Nakayama (1994) response or activation indicating the presence of a particular feature. Most models assume that responses are stronger at locations with high local feature contrast, consistent with neurophysiological data, e.g., the response of a red feature detector to a red object is stronger if the object is surrounded by green objects. The saliency map is obtained by taking a sum of bottom-up activations from the feature maps. The bottom-up activations are modulated by a top-down gain that specifies the contribution of a particular map to saliency in the current task and environment. Wolfe (1994) describes a heuristic algorithm for determining appropriate gains in a visual search task, where the goal is to detect a target object among distractor objects. Wolfe proposes that maps encoding features that discriminate between target and distractors have higher gains, and to be consistent with the data, he proposes limits on the magnitude of gain modulation and the number of gains that can be modulated. More recently, Wolfe et al. (2003) have been explicit in proposing optimization as a principle for setting gains given the task definition and stimulus environment. One aspect of optimizing attentional control involves configuring the attentional system to perform a given task; for example, in a visual search task for a red vertical target among green vertical and red horizontal distractors, the task definition should result in a higher gain for red and vertical feature maps than for other feature maps. However, there is a more subtle form of gain modulation, which depends on the statistics of display environments. For example, if green vertical distractors predominate, then red is a better discriminative cue than vertical; and if red horizontal distractors predominate, then vertical is a better discriminative cue than red. In this paper, we propose a model that encodes statistics of the environment in order to allow for optimization of attentional control to the structure of the environment. Our model is designed to address a key set of behavioral data, which we describe next. 1.1 Attentional priming phenomena Psychological studies involve a sequence of experimental trials that begin with a stimulus presentation and end with a response from the human participant. Typically, trial order is randomized, and the context preceding a trial is ignored. However, in sequential studies, performance is examined on one trial contingent on the past history of trials. These sequential studies explore how experience influences future performance. Consider a the sequential attentional task of Maljkovic and Nakayama (1994). On each trial, the stimulus display (Figure 2) consists of three notched diamonds, one a singleton in color—either green among red or red among green. The task is to report whether the singleton diamond, referred to as the target, is notched on the left or the right. The task is easy because the singleton pops out, i.e., the time to locate the singleton does not depend on the number of diamonds in the display. Nonetheless, the response time significantly depends on the sequence of trials leading up to the current trial: If the target is the same color on the cur- rent trial as on the previous trial, response time is roughly 100 ms faster than if the target is a different color on the current trial. Considering that response times are on the order of 700 ms, this effect, which we term attentional priming, is gigantic in the scheme of psychological phenomena. 2 ATTENTIONAL CONTROL AS ADAPTATION TO THE STATISTICS OF THE ENVIRONMENT We interpret the phenomenon of attentional priming via a particular perspective on attentional control, which can be summarized in two bullets. • The perceptual system dynamically constructs a probabilistic model of the environment based on its past experience. • Control parameters of the attentional system are tuned so as to optimize performance under the current environmental model. The primary focus of this paper is the environmental model, but we first discuss the nature of performance optimization. The role of attention is to make processing of some stimuli more efficient, and consequently, the processing of other stimuli less efficient. For example, if the gain on the red feature map is turned up, processing will be efficient for red items, but competition from red items will reduce the efficiency for green items. Thus, optimal control should tune the system for the most likely states of the world by minimizing an objective function such as: J(g) = ∑ P ( e )RT g ( e ) (1) e where g is a vector of top-down gains, e is an index over environmental states, P(.) is the probability of an environmental state, and RTg(.) is the expected response time—assuming a constant error rate—to the environmental state under gains g. Determining the optimal gains is a challenge because every gain setting will result in facilitation of responses to some environmental states but hindrance of responses to other states. The optimal control problem could be solved via direct reinforcement learning, but the rapidity of human learning makes this possibility unlikely: In a variety of experimental tasks, evidence suggests that adaptation to a new task or environment can occur in just one or two trials (e.g., Rogers & Monsell, 1996). Model-based reinforcement learning is an attractive alternative, because given a model, optimization can occur without further experience in the real world. Although the number of real-world trials necessary to achieve a given level of performance is comparable for direct and model-based reinforcement learning in stationary environments (Kearns & Singh, 1999), naturalistic environments can be viewed as highly nonstationary. In such a situation, the framework we suggest is well motivated: After each experience, the environment model is updated. The updated environmental model is then used to retune the attentional system. In this paper, we propose a particular model of the environment suitable for visual search tasks. Rather than explicitly modeling the optimization of attentional control by setting gains, we assume that the optimization process will serve to minimize Equation 1. Because any gain adjustment will facilitate performance in some environmental states and hinder performance in others, an optimized control system should obtain faster reaction times for more probable environmental states. This assumption allows us to explain experimental results in a minimal, parsimonious framework. 3 MODELING THE ENVIRONMENT Focusing on the domain of visual search, we characterize the environment in terms of a probability distribution over configurations of target and distractor features. We distinguish three classes of features: defining, reported, and irrelevant. To explain these terms, consider the task of searching a display of size varying, colored, notched diamonds (Figure 2), with the task of detecting the singleton in color and judging the notch location. Color is the defining feature, notch location is the reported feature, and size is an irrelevant feature. To simplify the exposition, we treat all features as having discrete values, an assumption which is true of the experimental tasks we model. We begin by considering displays containing a single target and a single distractor, and shortly generalize to multidistractor displays. We use the framework of Bayesian networks to characterize the environment. Each feature of the target and distractor is a discrete random variable, e.g., Tcolor for target color and Dnotch for the location of the notch on the distractor. The Bayes net encodes the probability distribution over environmental states; in our working example, this distribution is P(Tcolor, Tsize, Tnotch, Dcolor, Dsize, Dnotch). The structure of the Bayes net specifies the relationships among the features. The simplest model one could consider would be to treat the features as independent, illustrated in Figure 3a for singleton-color search task. The opposite extreme would be the full joint distribution, which could be represented by a look up table indexed by the six features, or by the cascading Bayes net architecture in Figure 3b. The architecture we propose, which we’ll refer to as the dominance model (Figure 3c), has an intermediate dependency structure, and expresses the joint distribution as: P(Tcolor)P(Dcolor |Tcolor)P(Tsize |Tcolor)P(Tnotch |Tcolor)P(Dsize |Dcolor)P(Dnotch |Tcolor). The structured model is constructed based on three rules. 1. The defining feature of the target is at the root of the tree. 2. The defining feature of the distractor is conditionally dependent on the defining feature of the target. We refer to this rule as dominance of the target over the distractor. 3. The reported and irrelevant features of target (distractor) are conditionally dependent on the defining feature of the target (distractor). We refer to this rule as dominance of the defining feature over nondefining features. As we will demonstrate, the dominance model produces a parsimonious account of a wide range of experimental data. 3.1 Updating the environment model The model’s parameters are the conditional distributions embodied in the links. In the example of Figure 3c with binary random variables, the model has 11 parameters. However, these parameters are determined by the environment: To be adaptive in nonstationary environments, the model must be updated following each experienced state. We propose a simple exponentially weighted averaging approach. For two variables V and W with observed values v and w on trial t, a conditional distribution, P t ( V = u W = w ) = δ uv , is (a) Tcolor Dcolor Tsize Tnotch (b) Tcolor Dcolor Dsize Tsize Dnotch Tnotch (c) Tcolor Dcolor Dsize Tsize Dsize Dnotch Tnotch Dnotch FIGURE 3. Three models of a visual-search environment with colored, notched, size-varying diamonds. (a) feature-independence model; (b) full-joint model; (c) dominance model. defined, where δ is the Kronecker delta. The distribution representing the environment E following trial t, denoted P t , is then updated as follows: E E P t ( V = u W = w ) = αP t – 1 ( V = u W = w ) + ( 1 – α )P t ( V = u W = w ) (2) for all u, where α is a memory constant. Note that no update is performed for values of W other than w. An analogous update is performed for unconditional distributions. E How the model is initialized—i.e., specifying P 0 —is irrelevant, because all experimental tasks that we model, participants begin the experiment with many dozens of practice trials. E Data is not collected during practice trials. Consequently, any transient effects of P 0 do E not impact the results. In our simulations, we begin with a uniform distribution for P 0 , and include practice trials as in the human studies. Thus far, we’ve assumed a single target and a single distractor. The experiments that we model involve multiple distractors. The simple extension we require to handle multiple distractors is to define a frequentist probability for each distractor feature V, P t ( V = v W = w ) = C vw ⁄ C w , where C vw is the count of co-occurrences of feature values v and w among the distractors, and C w is the count of w. Our model is extremely simple. Given a description of the visual search task and environment, the model has only a single degree of freedom, α . In all simulations, we fix α = 0.75 ; however, the choice of α does not qualitatively impact any result. 4 SIMULATIONS In this section, we show that the model can explain a range of data from four different experiments examining attentional priming. All experiments measure response times of participants. On each trial, the model can be used to obtain a probability of the display configuration (the environmental state) on that trial, given the history of trials to that point. Our critical assumption—as motivated earlier—is that response times monotonically decrease with increasing probability, indicating that visual information processing is better configured for more likely environmental states. The particular relationship we assume is that response times are linear in log probability. This assumption yields long response time tails, as are observed in all human studies. 4.1 Maljkovic and Nakayama (1994, Experiment 5) In this experiment, participants were asked to search for a singleton in color in a display of three red or green diamonds. Each diamond was notched on either the left or right side, and the task was to report the side of the notch on the color singleton. The well-practiced participants made very few errors. Reaction time (RT) was examined as a function of whether the target on a given trial is the same or different color as the target on trial n steps back or ahead. Figure 4 shows the results, with the human RTs in the left panel and the simulation log probabilities in the right panel. The horizontal axis represents n. Both graphs show the same outcome: repetition of target color facilitates performance. This influence lasts only for a half dozen trials, with an exponentially decreasing influence further into the past. In the model, this decreasing influence is due to the exponential decay of recent history (Equation 2). Figure 4 also shows that—as expected—the future has no influence on the current trial. 4.2 Maljkovic and Nakayama (1994, Experiment 8) In the previous experiment, it is impossible to determine whether facilitation is due to repetition of the target’s color or the distractor’s color, because the display contains only two colors, and therefore repetition of target color implies repetition of distractor color. To unconfound these two potential factors, an experiment like the previous one was con- ducted using four distinct colors, allowing one to examine the effect of repeating the target color while varying the distractor color, and vice versa. The sequence of trials was composed of subsequences of up-to-six consecutive trials with either the target or distractor color held constant while the other color was varied trial to trial. Following each subsequence, both target and distractors were changed. Figure 5 shows that for both humans and the simulation, performance improves toward an asymptote as the number of target and distractor repetitions increases; in the model, the asymptote is due to the probability of the repeated color in the environment model approaching 1.0. The performance improvement is greater for target than distractor repetition; in the model, this difference is due to the dominance of the defining feature of the target over the defining feature of the distractor. 4.3 Huang, Holcombe, and Pashler (2004, Experiment 1) Huang et al. (2004) and Hillstrom (2000) conducted studies to determine whether repetitions of one feature facilitate performance independently of repetitions of another feature. In the Huang et al. study, participants searched for a singleton in size in a display consisting of lines that were short and long, slanted left or right, and colored white or black. The reported feature was target slant. Slant, size, and color were uncorrelated. Huang et al. discovered that repeating an irrelevant feature (color or orientation) facilitated performance, but only when the defining feature (size) was repeated. As shown in Figure 6, the model replicates human performance, due to the dominance of the defining feature over the reported and irrelevant features. 4.4 Wolfe, Butcher, Lee, and Hyde (2003, Experiment 1) In an empirical tour-de-force, Wolfe et al. (2003) explored singleton search over a range of environments. The task is to detect the presence or absence of a singleton in displays conHuman data Different Color 600 Different Color 590 580 570 15 13 11 9 7 Past 5 3.2 3 Same Color 2.8 Same Color 560 550 Simulation 3.4 log(P(trial)) Reaction Time (msec) 610 3 1 +1 +3 +5 Future 2.6 +7 15 13 Relative Trial Number 11 9 7 Past 5 3 1 +1 +3 +5 Future +7 Relative Trial Number FIGURE 4. Experiment 5 of Maljkovic and Nakayama (1994): performance on a given trial conditional on the color of the target on a previous or subsequent trial. Human data is from subject KN. 650 6 Distractors Same 630 5.5 log(P(trial)) FIGURE 5. Experiment 8 of Maljkovic and Nakayama (1994). (left panel) human data, average of subjects KN and SS; (right panel) simulation Reaction Time (msec) 640 620 Target Same 610 5 Distractors Same 4.5 4 600 Target Same 3.5 590 3 580 1 2 3 4 5 1 6 4 5 6 4 1000 Size Alternate Size Alternate log(P(trial)) Reaction Time (msec) 3 4.2 1050 FIGURE 6. Experiment 1 of Huang, Holcombe, & Pashler (2004). (left panel) human data; (right panel) simulation 2 Order in Sequence Order in Sequence 950 3.8 3.6 3.4 900 3.2 Size Repeat 850 Size Repeat 3 Color Repeat Color Alternate Color Repeat Color Alternate sisting of colored (red or green), oriented (horizontal or vertical) lines. Target-absent trials were used primarily to ensure participants were searching the display. The experiment examined seven experimental conditions, which varied in the amount of uncertainty as to the target identity. The essential conditions, from least to most uncertainty, are: blocked (e.g., target always red vertical among green horizontals), mixed feature (e.g., target always a color singleton), mixed dimension (e.g., target either red or vertical), and fully mixed (target could be red, green, vertical, or horizontal). With this design, one can ascertain how uncertainty in the environment and in the target definition influence task difficulty. Because the defining feature in this experiment could be either color or orientation, we modeled the environment with two Bayes nets—one color dominant and one orientation dominant—and performed model averaging. A comparison of Figures 7a and 7b show a correspondence between human RTs and model predictions. Less uncertainty in the environment leads to more efficient performance. One interesting result from the model is its prediction that the mixed-feature condition is easier than the fully-mixed condition; that is, search is more efficient when the dimension (i.e., color vs. orientation) of the singleton is known, even though the model has no abstract representation of feature dimensions, only feature values. 4.5 Optimal adaptation constant In all simulations so far, we fixed the memory constant. From the human data, it is clear that memory for recent experience is relatively short lived, on the order of a half dozen trials (e.g., left panel of Figure 4). In this section we provide a rational argument for the short duration of memory in attentional control. Figure 7c shows mean negative log probability in each condition of the Wolfe et al. (2003) experiment, as a function of α . To assess these probabilities, for each experimental condition, the model was initialized so that all of the conditional distributions were uniform, and then a block of trials was run. Log probability for all trials in the block was averaged. The negative log probability (y axis of the Figure) is a measure of the model’s misprediction of the next trial in the sequence. For complex environments, such as the fully-mixed condition, a small memory constant is detrimental: With rapid memory decay, the effective history of trials is a high-variance sample of the distribution of environmental states. For simple environments, a large memory constant is detrimental: With slow memory decay, the model does not transition quickly from the initial environmental model to one that reflects the statistics of a new environment. Thus, the memory constant is constrained by being large enough that the environment model can hold on to sufficient history to represent complex environments, and by being small enough that the model adapts quickly to novel environments. If the conditions in Wolfe et al. give some indication of the range of naturalistic environments an agent encounters, we have a rational account of why attentional priming is so short lived. Whether priming lasts 2 trials or 20, the surprising empirical result is that it does not last 200 or 2000 trials. Our rational argument provides a rough insight into this finding. (a) fully mixed mixed feature mixed dimension blocked 460 (c) Simulation fully mixed mixed feature mixed dimension blocked 4 5 420 log(P(trial)) 440 2 Blocked Red or Vertical Blocked Red and Vertical Mixed Feature Mixed Dimension Fully Mixed 4 3 log(P(trial)) reaction time (msec) (b) Human Data 480 3 2 1 400 1 380 0 360 0 red or vert red and vert target type red or vert red and vert target type 0 0.5 0.8 0.9 0.95 0.98 Memory Constant FIGURE 7. (a) Human data for Wolfe et al. (2003), Experiment 1; (b) simulation; (c) misprediction of model (i.e., lower y value = better) as a function of α for five experimental condition 5 DISCUSSION The psychological literature contains two opposing accounts of attentional priming and its relation to attentional control. Huang et al. (2004) and Hillstrom (2000) propose an episodic account in which a distinct memory trace—representing the complete configuration of features in the display—is laid down for each trial, and priming depends on configural similarity of the current trial to previous trials. Alternatively, Maljkovic and Nakayama (1994) and Wolfe et al. (2003) propose a feature-strengthening account in which detection of a feature on one trial increases its ability to attract attention on subsequent trials, and priming is proportional to the number of overlapping features from one trial to the next. The episodic account corresponds roughly to the full joint model (Figure 3b), and the feature-strengthening account corresponds roughly to the independence model (Figure 3a). Neither account is adequate to explain the range of data we presented. However, an intermediate account, the dominance model (Figure 3c), is not only sufficient, but it offers a parsimonious, rational explanation. Beyond the model’s basic assumptions, it has only one free parameter, and can explain results from diverse experimental paradigms. The model makes a further theoretical contribution. Wolfe et al. distinguish the environments in their experiment in terms of the amount of top-down control available, implying that different mechanisms might be operating in different environments. However, in our account, top-down control is not some substance distributed in different amounts depending on the nature of the environment. Our account treats all environments uniformly, relying on attentional control to adapt to the environment at hand. We conclude with two limitations of the present work. First, our account presumes a particular network architecture, instead of a more elegant Bayesian approach that specifies priors over architectures, and performs automatic model selection via the sequence of trials. We did explore such a Bayesian approach, but it was unable to explain the data. Second, at least one finding in the literature is problematic for the model. Hillstrom (2000) occasionally finds that RTs slow when an irrelevant target feature is repeated but the defining target feature is not. However, because this effect is observed only in some experiments, it is likely that any model would require elaboration to explain the variability. ACKNOWLEDGEMENTS We thank Jeremy Wolfe for providing the raw data from his experiment for reanalysis. This research was funded by NSF BCS Award 0339103. REFERENCES Huang, L, Holcombe, A. O., & Pashler, H. (2004). Repetition priming in visual search: Episodic retrieval, not feature priming. Memory & Cognition, 32, 12–20. Hillstrom, A. P. (2000). Repetition effects in visual search. Perception & Psychophysics, 62, 800-817. Itti, L., Koch, C., & Niebur, E. (1998). A model of saliency-based visual attention for rapid scene analysis. IEEE Trans. Pattern Analysis & Machine Intelligence, 20, 1254–1259. Kearns, M., & Singh, S. (1999). Finite-sample convergence rates for Q-learning and indirect algorithms. In Advances in Neural Information Processing Systems 11 (pp. 996–1002). Cambridge, MA: MIT Press. Koch, C. and Ullman, S. (1985). Shifts in selective visual attention: towards the underlying neural circuitry. Human Neurobiology, 4, 219–227. Maljkovic, V., & Nakayama, K. (1994). Priming of pop-out: I. Role of features. Mem. & Cognition, 22, 657-672. Mozer, M. C. (1991). The perception of multiple objects: A connectionist approach. Cambridge, MA: MIT Press. Rogers, R. D., & Monsell, S. (1995). The cost of a predictable switch between simple cognitive tasks. Journal of Experimental Psychology: General, 124, 207–231. Wolfe, J.M. (1994). Guided Search 2.0: A Revised Model of Visual Search. Psych. Bull. & Rev., 1, 202–238. Wolfe, J. S., Butcher, S. J., Lee, C., & Hyde, M. (2003). Changing your mind: on the contributions of top-down and bottom-up guidance in visual search for feature singletons. Journal of Exptl. Psychology: Human Perception & Performance, 29, 483-502.
2 0.23149174 149 nips-2005-Optimal cue selection strategy
Author: Vidhya Navalpakkam, Laurent Itti
Abstract: Survival in the natural world demands the selection of relevant visual cues to rapidly and reliably guide attention towards prey and predators in cluttered environments. We investigate whether our visual system selects cues that guide search in an optimal manner. We formally obtain the optimal cue selection strategy by maximizing the signal to noise ratio (SN R) between a search target and surrounding distractors. This optimal strategy successfully accounts for several phenomena in visual search behavior, including the effect of target-distractor discriminability, uncertainty in target’s features, distractor heterogeneity, and linear separability. Furthermore, the theory generates a new prediction, which we verify through psychophysical experiments with human subjects. Our results provide direct experimental evidence that humans select visual cues so as to maximize SN R between the targets and surrounding clutter.
3 0.17938362 141 nips-2005-Norepinephrine and Neural Interrupts
Author: Peter Dayan, Angela J. Yu
Abstract: Experimental data indicate that norepinephrine is critically involved in aspects of vigilance and attention. Previously, we considered the function of this neuromodulatory system on a time scale of minutes and longer, and suggested that it signals global uncertainty arising from gross changes in environmental contingencies. However, norepinephrine is also known to be activated phasically by familiar stimuli in welllearned tasks. Here, we extend our uncertainty-based treatment of norepinephrine to this phasic mode, proposing that it is involved in the detection and reaction to state uncertainty within a task. This role of norepinephrine can be understood through the metaphor of neural interrupts. 1
4 0.16245767 169 nips-2005-Saliency Based on Information Maximization
Author: Neil Bruce, John Tsotsos
Abstract: A model of bottom-up overt attention is proposed based on the principle of maximizing information sampled from a scene. The proposed operation is based on Shannon's self-information measure and is achieved in a neural circuit, which is demonstrated as having close ties with the circuitry existent in the primate visual cortex. It is further shown that the proposed saliency measure may be extended to address issues that currently elude explanation in the domain of saliency based models. Resu lts on natural images are compared with experimental eye tracking data revealing the efficacy of the model in predicting the deployment of overt attention as compared with existing efforts.
5 0.15723568 5 nips-2005-A Computational Model of Eye Movements during Object Class Detection
Author: Wei Zhang, Hyejin Yang, Dimitris Samaras, Gregory J. Zelinsky
Abstract: We present a computational model of human eye movements in an object class detection task. The model combines state-of-the-art computer vision object class detection methods (SIFT features trained using AdaBoost) with a biologically plausible model of human eye movement to produce a sequence of simulated fixations, culminating with the acquisition of a target. We validated the model by comparing its behavior to the behavior of human observers performing the identical object class detection task (looking for a teddy bear among visually complex nontarget objects). We found considerable agreement between the model and human data in multiple eye movement measures, including number of fixations, cumulative probability of fixating the target, and scanpath distance.
6 0.09325356 193 nips-2005-The Role of Top-down and Bottom-up Processes in Guiding Eye Movements during Visual Search
7 0.092505634 140 nips-2005-Nonparametric inference of prior probabilities from Bayes-optimal behavior
8 0.082556829 67 nips-2005-Extracting Dynamical Structure Embedded in Neural Activity
9 0.078626521 7 nips-2005-A Cortically-Plausible Inverse Problem Solving Method Applied to Recognizing Static and Kinematic 3D Objects
10 0.076918781 26 nips-2005-An exploration-exploitation model based on norepinepherine and dopamine activity
11 0.064777918 187 nips-2005-Temporal Abstraction in Temporal-difference Networks
12 0.063317202 36 nips-2005-Bayesian models of human action understanding
13 0.063312761 156 nips-2005-Prediction and Change Detection
14 0.063173622 132 nips-2005-Nearest Neighbor Based Feature Selection for Regression and its Application to Neural Activity
15 0.061500035 34 nips-2005-Bayesian Surprise Attracts Human Attention
16 0.059705526 173 nips-2005-Sensory Adaptation within a Bayesian Framework for Perception
17 0.058543701 94 nips-2005-Identifying Distributed Object Representations in Human Extrastriate Visual Cortex
18 0.055848084 143 nips-2005-Off-Road Obstacle Avoidance through End-to-End Learning
19 0.054816049 109 nips-2005-Learning Cue-Invariant Visual Responses
20 0.052461721 3 nips-2005-A Bayesian Framework for Tilt Perception and Confidence
topicId topicWeight
[(0, 0.185), (1, -0.1), (2, 0.073), (3, 0.207), (4, 0.007), (5, 0.174), (6, 0.001), (7, -0.088), (8, -0.208), (9, 0.097), (10, 0.127), (11, -0.03), (12, -0.08), (13, 0.088), (14, -0.082), (15, -0.032), (16, -0.041), (17, 0.078), (18, 0.056), (19, -0.154), (20, 0.017), (21, -0.133), (22, 0.003), (23, 0.042), (24, 0.059), (25, -0.054), (26, 0.015), (27, 0.077), (28, 0.012), (29, 0.093), (30, 0.044), (31, 0.005), (32, 0.004), (33, -0.01), (34, -0.092), (35, 0.049), (36, -0.084), (37, -0.019), (38, 0.057), (39, -0.017), (40, 0.014), (41, -0.022), (42, 0.128), (43, 0.031), (44, 0.08), (45, -0.087), (46, -0.078), (47, 0.094), (48, -0.011), (49, 0.042)]
simIndex simValue paperId paperTitle
same-paper 1 0.96805048 194 nips-2005-Top-Down Control of Visual Attention: A Rational Account
Author: Michael Shettel, Shaun Vecera, Michael C. Mozer
Abstract: Theories of visual attention commonly posit that early parallel processes extract conspicuous features such as color contrast and motion from the visual field. These features are then combined into a saliency map, and attention is directed to the most salient regions first. Top-down attentional control is achieved by modulating the contribution of different feature types to the saliency map. A key source of data concerning attentional control comes from behavioral studies in which the effect of recent experience is examined as individuals repeatedly perform a perceptual discrimination task (e.g., “what shape is the odd-colored object?”). The robust finding is that repetition of features of recent trials (e.g., target color) facilitates performance. We view this facilitation as an adaptation to the statistical structure of the environment. We propose a probabilistic model of the environment that is updated after each trial. Under the assumption that attentional control operates so as to make performance more efficient for more likely environmental states, we obtain parsimonious explanations for data from four different experiments. Further, our model provides a rational explanation for why the influence of past experience on attentional control is short lived. 1 INTRODUCTION The brain does not have the computational capacity to fully process the massive quantity of information provided by the eyes. Selective attention operates to filter the spatiotemporal stream to a manageable quantity. Key to understanding the nature of attention is discovering the algorithm governing selection, i.e., understanding what information will be selected and what will be suppressed. Selection is influenced by attributes of the spatiotemporal stream, often referred to as bottom-up contributions to attention. For example, attention is drawn to abrupt onsets, motion, and regions of high contrast in brightness and color. Most theories of attention posit that some visual information processing is performed preattentively and in parallel across the visual field. This processing extracts primitive visual features such as color and motion, which provide the bottom-up cues for attentional guidance. However, attention is not driven willy nilly by these cues. The deployment of attention can be modulated by task instructions, current goals, and domain knowledge, collectively referred to as top-down contributions to attention. How do bottom-up and top-down contributions to attention interact? Most psychologically and neurobiologically motivated models propose a very similar architecture in which information from bottom-up and top-down sources combines in a saliency (or activation) map (e.g., Itti et al., 1998; Koch & Ullman, 1985; Mozer, 1991; Wolfe, 1994). The saliency map indicates, for each location in the visual field, the relative importance of that location. Attention is drawn to the most salient locations first. Figure 1 sketches the basic architecture that incorporates bottom-up and top-down contributions to the saliency map. The visual image is analyzed to extract maps of primitive features such as color and orientation. Associated with each location in a map is a scalar visual image horizontal primitive feature maps vertical green top-down gains red saliency map FIGURE 1. An attentional saliency map constructed from bottom-up and top-down information bottom-up activations FIGURE 2. Sample display from Experiment 1 of Maljkovic and Nakayama (1994) response or activation indicating the presence of a particular feature. Most models assume that responses are stronger at locations with high local feature contrast, consistent with neurophysiological data, e.g., the response of a red feature detector to a red object is stronger if the object is surrounded by green objects. The saliency map is obtained by taking a sum of bottom-up activations from the feature maps. The bottom-up activations are modulated by a top-down gain that specifies the contribution of a particular map to saliency in the current task and environment. Wolfe (1994) describes a heuristic algorithm for determining appropriate gains in a visual search task, where the goal is to detect a target object among distractor objects. Wolfe proposes that maps encoding features that discriminate between target and distractors have higher gains, and to be consistent with the data, he proposes limits on the magnitude of gain modulation and the number of gains that can be modulated. More recently, Wolfe et al. (2003) have been explicit in proposing optimization as a principle for setting gains given the task definition and stimulus environment. One aspect of optimizing attentional control involves configuring the attentional system to perform a given task; for example, in a visual search task for a red vertical target among green vertical and red horizontal distractors, the task definition should result in a higher gain for red and vertical feature maps than for other feature maps. However, there is a more subtle form of gain modulation, which depends on the statistics of display environments. For example, if green vertical distractors predominate, then red is a better discriminative cue than vertical; and if red horizontal distractors predominate, then vertical is a better discriminative cue than red. In this paper, we propose a model that encodes statistics of the environment in order to allow for optimization of attentional control to the structure of the environment. Our model is designed to address a key set of behavioral data, which we describe next. 1.1 Attentional priming phenomena Psychological studies involve a sequence of experimental trials that begin with a stimulus presentation and end with a response from the human participant. Typically, trial order is randomized, and the context preceding a trial is ignored. However, in sequential studies, performance is examined on one trial contingent on the past history of trials. These sequential studies explore how experience influences future performance. Consider a the sequential attentional task of Maljkovic and Nakayama (1994). On each trial, the stimulus display (Figure 2) consists of three notched diamonds, one a singleton in color—either green among red or red among green. The task is to report whether the singleton diamond, referred to as the target, is notched on the left or the right. The task is easy because the singleton pops out, i.e., the time to locate the singleton does not depend on the number of diamonds in the display. Nonetheless, the response time significantly depends on the sequence of trials leading up to the current trial: If the target is the same color on the cur- rent trial as on the previous trial, response time is roughly 100 ms faster than if the target is a different color on the current trial. Considering that response times are on the order of 700 ms, this effect, which we term attentional priming, is gigantic in the scheme of psychological phenomena. 2 ATTENTIONAL CONTROL AS ADAPTATION TO THE STATISTICS OF THE ENVIRONMENT We interpret the phenomenon of attentional priming via a particular perspective on attentional control, which can be summarized in two bullets. • The perceptual system dynamically constructs a probabilistic model of the environment based on its past experience. • Control parameters of the attentional system are tuned so as to optimize performance under the current environmental model. The primary focus of this paper is the environmental model, but we first discuss the nature of performance optimization. The role of attention is to make processing of some stimuli more efficient, and consequently, the processing of other stimuli less efficient. For example, if the gain on the red feature map is turned up, processing will be efficient for red items, but competition from red items will reduce the efficiency for green items. Thus, optimal control should tune the system for the most likely states of the world by minimizing an objective function such as: J(g) = ∑ P ( e )RT g ( e ) (1) e where g is a vector of top-down gains, e is an index over environmental states, P(.) is the probability of an environmental state, and RTg(.) is the expected response time—assuming a constant error rate—to the environmental state under gains g. Determining the optimal gains is a challenge because every gain setting will result in facilitation of responses to some environmental states but hindrance of responses to other states. The optimal control problem could be solved via direct reinforcement learning, but the rapidity of human learning makes this possibility unlikely: In a variety of experimental tasks, evidence suggests that adaptation to a new task or environment can occur in just one or two trials (e.g., Rogers & Monsell, 1996). Model-based reinforcement learning is an attractive alternative, because given a model, optimization can occur without further experience in the real world. Although the number of real-world trials necessary to achieve a given level of performance is comparable for direct and model-based reinforcement learning in stationary environments (Kearns & Singh, 1999), naturalistic environments can be viewed as highly nonstationary. In such a situation, the framework we suggest is well motivated: After each experience, the environment model is updated. The updated environmental model is then used to retune the attentional system. In this paper, we propose a particular model of the environment suitable for visual search tasks. Rather than explicitly modeling the optimization of attentional control by setting gains, we assume that the optimization process will serve to minimize Equation 1. Because any gain adjustment will facilitate performance in some environmental states and hinder performance in others, an optimized control system should obtain faster reaction times for more probable environmental states. This assumption allows us to explain experimental results in a minimal, parsimonious framework. 3 MODELING THE ENVIRONMENT Focusing on the domain of visual search, we characterize the environment in terms of a probability distribution over configurations of target and distractor features. We distinguish three classes of features: defining, reported, and irrelevant. To explain these terms, consider the task of searching a display of size varying, colored, notched diamonds (Figure 2), with the task of detecting the singleton in color and judging the notch location. Color is the defining feature, notch location is the reported feature, and size is an irrelevant feature. To simplify the exposition, we treat all features as having discrete values, an assumption which is true of the experimental tasks we model. We begin by considering displays containing a single target and a single distractor, and shortly generalize to multidistractor displays. We use the framework of Bayesian networks to characterize the environment. Each feature of the target and distractor is a discrete random variable, e.g., Tcolor for target color and Dnotch for the location of the notch on the distractor. The Bayes net encodes the probability distribution over environmental states; in our working example, this distribution is P(Tcolor, Tsize, Tnotch, Dcolor, Dsize, Dnotch). The structure of the Bayes net specifies the relationships among the features. The simplest model one could consider would be to treat the features as independent, illustrated in Figure 3a for singleton-color search task. The opposite extreme would be the full joint distribution, which could be represented by a look up table indexed by the six features, or by the cascading Bayes net architecture in Figure 3b. The architecture we propose, which we’ll refer to as the dominance model (Figure 3c), has an intermediate dependency structure, and expresses the joint distribution as: P(Tcolor)P(Dcolor |Tcolor)P(Tsize |Tcolor)P(Tnotch |Tcolor)P(Dsize |Dcolor)P(Dnotch |Tcolor). The structured model is constructed based on three rules. 1. The defining feature of the target is at the root of the tree. 2. The defining feature of the distractor is conditionally dependent on the defining feature of the target. We refer to this rule as dominance of the target over the distractor. 3. The reported and irrelevant features of target (distractor) are conditionally dependent on the defining feature of the target (distractor). We refer to this rule as dominance of the defining feature over nondefining features. As we will demonstrate, the dominance model produces a parsimonious account of a wide range of experimental data. 3.1 Updating the environment model The model’s parameters are the conditional distributions embodied in the links. In the example of Figure 3c with binary random variables, the model has 11 parameters. However, these parameters are determined by the environment: To be adaptive in nonstationary environments, the model must be updated following each experienced state. We propose a simple exponentially weighted averaging approach. For two variables V and W with observed values v and w on trial t, a conditional distribution, P t ( V = u W = w ) = δ uv , is (a) Tcolor Dcolor Tsize Tnotch (b) Tcolor Dcolor Dsize Tsize Dnotch Tnotch (c) Tcolor Dcolor Dsize Tsize Dsize Dnotch Tnotch Dnotch FIGURE 3. Three models of a visual-search environment with colored, notched, size-varying diamonds. (a) feature-independence model; (b) full-joint model; (c) dominance model. defined, where δ is the Kronecker delta. The distribution representing the environment E following trial t, denoted P t , is then updated as follows: E E P t ( V = u W = w ) = αP t – 1 ( V = u W = w ) + ( 1 – α )P t ( V = u W = w ) (2) for all u, where α is a memory constant. Note that no update is performed for values of W other than w. An analogous update is performed for unconditional distributions. E How the model is initialized—i.e., specifying P 0 —is irrelevant, because all experimental tasks that we model, participants begin the experiment with many dozens of practice trials. E Data is not collected during practice trials. Consequently, any transient effects of P 0 do E not impact the results. In our simulations, we begin with a uniform distribution for P 0 , and include practice trials as in the human studies. Thus far, we’ve assumed a single target and a single distractor. The experiments that we model involve multiple distractors. The simple extension we require to handle multiple distractors is to define a frequentist probability for each distractor feature V, P t ( V = v W = w ) = C vw ⁄ C w , where C vw is the count of co-occurrences of feature values v and w among the distractors, and C w is the count of w. Our model is extremely simple. Given a description of the visual search task and environment, the model has only a single degree of freedom, α . In all simulations, we fix α = 0.75 ; however, the choice of α does not qualitatively impact any result. 4 SIMULATIONS In this section, we show that the model can explain a range of data from four different experiments examining attentional priming. All experiments measure response times of participants. On each trial, the model can be used to obtain a probability of the display configuration (the environmental state) on that trial, given the history of trials to that point. Our critical assumption—as motivated earlier—is that response times monotonically decrease with increasing probability, indicating that visual information processing is better configured for more likely environmental states. The particular relationship we assume is that response times are linear in log probability. This assumption yields long response time tails, as are observed in all human studies. 4.1 Maljkovic and Nakayama (1994, Experiment 5) In this experiment, participants were asked to search for a singleton in color in a display of three red or green diamonds. Each diamond was notched on either the left or right side, and the task was to report the side of the notch on the color singleton. The well-practiced participants made very few errors. Reaction time (RT) was examined as a function of whether the target on a given trial is the same or different color as the target on trial n steps back or ahead. Figure 4 shows the results, with the human RTs in the left panel and the simulation log probabilities in the right panel. The horizontal axis represents n. Both graphs show the same outcome: repetition of target color facilitates performance. This influence lasts only for a half dozen trials, with an exponentially decreasing influence further into the past. In the model, this decreasing influence is due to the exponential decay of recent history (Equation 2). Figure 4 also shows that—as expected—the future has no influence on the current trial. 4.2 Maljkovic and Nakayama (1994, Experiment 8) In the previous experiment, it is impossible to determine whether facilitation is due to repetition of the target’s color or the distractor’s color, because the display contains only two colors, and therefore repetition of target color implies repetition of distractor color. To unconfound these two potential factors, an experiment like the previous one was con- ducted using four distinct colors, allowing one to examine the effect of repeating the target color while varying the distractor color, and vice versa. The sequence of trials was composed of subsequences of up-to-six consecutive trials with either the target or distractor color held constant while the other color was varied trial to trial. Following each subsequence, both target and distractors were changed. Figure 5 shows that for both humans and the simulation, performance improves toward an asymptote as the number of target and distractor repetitions increases; in the model, the asymptote is due to the probability of the repeated color in the environment model approaching 1.0. The performance improvement is greater for target than distractor repetition; in the model, this difference is due to the dominance of the defining feature of the target over the defining feature of the distractor. 4.3 Huang, Holcombe, and Pashler (2004, Experiment 1) Huang et al. (2004) and Hillstrom (2000) conducted studies to determine whether repetitions of one feature facilitate performance independently of repetitions of another feature. In the Huang et al. study, participants searched for a singleton in size in a display consisting of lines that were short and long, slanted left or right, and colored white or black. The reported feature was target slant. Slant, size, and color were uncorrelated. Huang et al. discovered that repeating an irrelevant feature (color or orientation) facilitated performance, but only when the defining feature (size) was repeated. As shown in Figure 6, the model replicates human performance, due to the dominance of the defining feature over the reported and irrelevant features. 4.4 Wolfe, Butcher, Lee, and Hyde (2003, Experiment 1) In an empirical tour-de-force, Wolfe et al. (2003) explored singleton search over a range of environments. The task is to detect the presence or absence of a singleton in displays conHuman data Different Color 600 Different Color 590 580 570 15 13 11 9 7 Past 5 3.2 3 Same Color 2.8 Same Color 560 550 Simulation 3.4 log(P(trial)) Reaction Time (msec) 610 3 1 +1 +3 +5 Future 2.6 +7 15 13 Relative Trial Number 11 9 7 Past 5 3 1 +1 +3 +5 Future +7 Relative Trial Number FIGURE 4. Experiment 5 of Maljkovic and Nakayama (1994): performance on a given trial conditional on the color of the target on a previous or subsequent trial. Human data is from subject KN. 650 6 Distractors Same 630 5.5 log(P(trial)) FIGURE 5. Experiment 8 of Maljkovic and Nakayama (1994). (left panel) human data, average of subjects KN and SS; (right panel) simulation Reaction Time (msec) 640 620 Target Same 610 5 Distractors Same 4.5 4 600 Target Same 3.5 590 3 580 1 2 3 4 5 1 6 4 5 6 4 1000 Size Alternate Size Alternate log(P(trial)) Reaction Time (msec) 3 4.2 1050 FIGURE 6. Experiment 1 of Huang, Holcombe, & Pashler (2004). (left panel) human data; (right panel) simulation 2 Order in Sequence Order in Sequence 950 3.8 3.6 3.4 900 3.2 Size Repeat 850 Size Repeat 3 Color Repeat Color Alternate Color Repeat Color Alternate sisting of colored (red or green), oriented (horizontal or vertical) lines. Target-absent trials were used primarily to ensure participants were searching the display. The experiment examined seven experimental conditions, which varied in the amount of uncertainty as to the target identity. The essential conditions, from least to most uncertainty, are: blocked (e.g., target always red vertical among green horizontals), mixed feature (e.g., target always a color singleton), mixed dimension (e.g., target either red or vertical), and fully mixed (target could be red, green, vertical, or horizontal). With this design, one can ascertain how uncertainty in the environment and in the target definition influence task difficulty. Because the defining feature in this experiment could be either color or orientation, we modeled the environment with two Bayes nets—one color dominant and one orientation dominant—and performed model averaging. A comparison of Figures 7a and 7b show a correspondence between human RTs and model predictions. Less uncertainty in the environment leads to more efficient performance. One interesting result from the model is its prediction that the mixed-feature condition is easier than the fully-mixed condition; that is, search is more efficient when the dimension (i.e., color vs. orientation) of the singleton is known, even though the model has no abstract representation of feature dimensions, only feature values. 4.5 Optimal adaptation constant In all simulations so far, we fixed the memory constant. From the human data, it is clear that memory for recent experience is relatively short lived, on the order of a half dozen trials (e.g., left panel of Figure 4). In this section we provide a rational argument for the short duration of memory in attentional control. Figure 7c shows mean negative log probability in each condition of the Wolfe et al. (2003) experiment, as a function of α . To assess these probabilities, for each experimental condition, the model was initialized so that all of the conditional distributions were uniform, and then a block of trials was run. Log probability for all trials in the block was averaged. The negative log probability (y axis of the Figure) is a measure of the model’s misprediction of the next trial in the sequence. For complex environments, such as the fully-mixed condition, a small memory constant is detrimental: With rapid memory decay, the effective history of trials is a high-variance sample of the distribution of environmental states. For simple environments, a large memory constant is detrimental: With slow memory decay, the model does not transition quickly from the initial environmental model to one that reflects the statistics of a new environment. Thus, the memory constant is constrained by being large enough that the environment model can hold on to sufficient history to represent complex environments, and by being small enough that the model adapts quickly to novel environments. If the conditions in Wolfe et al. give some indication of the range of naturalistic environments an agent encounters, we have a rational account of why attentional priming is so short lived. Whether priming lasts 2 trials or 20, the surprising empirical result is that it does not last 200 or 2000 trials. Our rational argument provides a rough insight into this finding. (a) fully mixed mixed feature mixed dimension blocked 460 (c) Simulation fully mixed mixed feature mixed dimension blocked 4 5 420 log(P(trial)) 440 2 Blocked Red or Vertical Blocked Red and Vertical Mixed Feature Mixed Dimension Fully Mixed 4 3 log(P(trial)) reaction time (msec) (b) Human Data 480 3 2 1 400 1 380 0 360 0 red or vert red and vert target type red or vert red and vert target type 0 0.5 0.8 0.9 0.95 0.98 Memory Constant FIGURE 7. (a) Human data for Wolfe et al. (2003), Experiment 1; (b) simulation; (c) misprediction of model (i.e., lower y value = better) as a function of α for five experimental condition 5 DISCUSSION The psychological literature contains two opposing accounts of attentional priming and its relation to attentional control. Huang et al. (2004) and Hillstrom (2000) propose an episodic account in which a distinct memory trace—representing the complete configuration of features in the display—is laid down for each trial, and priming depends on configural similarity of the current trial to previous trials. Alternatively, Maljkovic and Nakayama (1994) and Wolfe et al. (2003) propose a feature-strengthening account in which detection of a feature on one trial increases its ability to attract attention on subsequent trials, and priming is proportional to the number of overlapping features from one trial to the next. The episodic account corresponds roughly to the full joint model (Figure 3b), and the feature-strengthening account corresponds roughly to the independence model (Figure 3a). Neither account is adequate to explain the range of data we presented. However, an intermediate account, the dominance model (Figure 3c), is not only sufficient, but it offers a parsimonious, rational explanation. Beyond the model’s basic assumptions, it has only one free parameter, and can explain results from diverse experimental paradigms. The model makes a further theoretical contribution. Wolfe et al. distinguish the environments in their experiment in terms of the amount of top-down control available, implying that different mechanisms might be operating in different environments. However, in our account, top-down control is not some substance distributed in different amounts depending on the nature of the environment. Our account treats all environments uniformly, relying on attentional control to adapt to the environment at hand. We conclude with two limitations of the present work. First, our account presumes a particular network architecture, instead of a more elegant Bayesian approach that specifies priors over architectures, and performs automatic model selection via the sequence of trials. We did explore such a Bayesian approach, but it was unable to explain the data. Second, at least one finding in the literature is problematic for the model. Hillstrom (2000) occasionally finds that RTs slow when an irrelevant target feature is repeated but the defining target feature is not. However, because this effect is observed only in some experiments, it is likely that any model would require elaboration to explain the variability. ACKNOWLEDGEMENTS We thank Jeremy Wolfe for providing the raw data from his experiment for reanalysis. This research was funded by NSF BCS Award 0339103. REFERENCES Huang, L, Holcombe, A. O., & Pashler, H. (2004). Repetition priming in visual search: Episodic retrieval, not feature priming. Memory & Cognition, 32, 12–20. Hillstrom, A. P. (2000). Repetition effects in visual search. Perception & Psychophysics, 62, 800-817. Itti, L., Koch, C., & Niebur, E. (1998). A model of saliency-based visual attention for rapid scene analysis. IEEE Trans. Pattern Analysis & Machine Intelligence, 20, 1254–1259. Kearns, M., & Singh, S. (1999). Finite-sample convergence rates for Q-learning and indirect algorithms. In Advances in Neural Information Processing Systems 11 (pp. 996–1002). Cambridge, MA: MIT Press. Koch, C. and Ullman, S. (1985). Shifts in selective visual attention: towards the underlying neural circuitry. Human Neurobiology, 4, 219–227. Maljkovic, V., & Nakayama, K. (1994). Priming of pop-out: I. Role of features. Mem. & Cognition, 22, 657-672. Mozer, M. C. (1991). The perception of multiple objects: A connectionist approach. Cambridge, MA: MIT Press. Rogers, R. D., & Monsell, S. (1995). The cost of a predictable switch between simple cognitive tasks. Journal of Experimental Psychology: General, 124, 207–231. Wolfe, J.M. (1994). Guided Search 2.0: A Revised Model of Visual Search. Psych. Bull. & Rev., 1, 202–238. Wolfe, J. S., Butcher, S. J., Lee, C., & Hyde, M. (2003). Changing your mind: on the contributions of top-down and bottom-up guidance in visual search for feature singletons. Journal of Exptl. Psychology: Human Perception & Performance, 29, 483-502.
2 0.87317878 149 nips-2005-Optimal cue selection strategy
Author: Vidhya Navalpakkam, Laurent Itti
Abstract: Survival in the natural world demands the selection of relevant visual cues to rapidly and reliably guide attention towards prey and predators in cluttered environments. We investigate whether our visual system selects cues that guide search in an optimal manner. We formally obtain the optimal cue selection strategy by maximizing the signal to noise ratio (SN R) between a search target and surrounding distractors. This optimal strategy successfully accounts for several phenomena in visual search behavior, including the effect of target-distractor discriminability, uncertainty in target’s features, distractor heterogeneity, and linear separability. Furthermore, the theory generates a new prediction, which we verify through psychophysical experiments with human subjects. Our results provide direct experimental evidence that humans select visual cues so as to maximize SN R between the targets and surrounding clutter.
3 0.70443755 141 nips-2005-Norepinephrine and Neural Interrupts
Author: Peter Dayan, Angela J. Yu
Abstract: Experimental data indicate that norepinephrine is critically involved in aspects of vigilance and attention. Previously, we considered the function of this neuromodulatory system on a time scale of minutes and longer, and suggested that it signals global uncertainty arising from gross changes in environmental contingencies. However, norepinephrine is also known to be activated phasically by familiar stimuli in welllearned tasks. Here, we extend our uncertainty-based treatment of norepinephrine to this phasic mode, proposing that it is involved in the detection and reaction to state uncertainty within a task. This role of norepinephrine can be understood through the metaphor of neural interrupts. 1
4 0.61016935 193 nips-2005-The Role of Top-down and Bottom-up Processes in Guiding Eye Movements during Visual Search
Author: Gregory Zelinsky, Wei Zhang, Bing Yu, Xin Chen, Dimitris Samaras
Abstract: To investigate how top-down (TD) and bottom-up (BU) information is weighted in the guidance of human search behavior, we manipulated the proportions of BU and TD components in a saliency-based model. The model is biologically plausible and implements an artificial retina and a neuronal population code. The BU component is based on featurecontrast. The TD component is defined by a feature-template match to a stored target representation. We compared the model’s behavior at different mixtures of TD and BU components to the eye movement behavior of human observers performing the identical search task. We found that a purely TD model provides a much closer match to human behavior than any mixture model using BU information. Only when biological constraints are removed (e.g., eliminating the retina) did a BU/TD mixture model begin to approximate human behavior.
5 0.58928072 169 nips-2005-Saliency Based on Information Maximization
Author: Neil Bruce, John Tsotsos
Abstract: A model of bottom-up overt attention is proposed based on the principle of maximizing information sampled from a scene. The proposed operation is based on Shannon's self-information measure and is achieved in a neural circuit, which is demonstrated as having close ties with the circuitry existent in the primate visual cortex. It is further shown that the proposed saliency measure may be extended to address issues that currently elude explanation in the domain of saliency based models. Resu lts on natural images are compared with experimental eye tracking data revealing the efficacy of the model in predicting the deployment of overt attention as compared with existing efforts.
6 0.53167337 3 nips-2005-A Bayesian Framework for Tilt Perception and Confidence
7 0.52420026 5 nips-2005-A Computational Model of Eye Movements during Object Class Detection
8 0.44343463 176 nips-2005-Silicon growth cones map silicon retina
9 0.39719093 26 nips-2005-An exploration-exploitation model based on norepinepherine and dopamine activity
10 0.38835311 132 nips-2005-Nearest Neighbor Based Feature Selection for Regression and its Application to Neural Activity
11 0.37754366 146 nips-2005-On the Accuracy of Bounded Rationality: How Far from Optimal Is Fast and Frugal?
12 0.34558931 34 nips-2005-Bayesian Surprise Attracts Human Attention
13 0.33242837 140 nips-2005-Nonparametric inference of prior probabilities from Bayes-optimal behavior
14 0.32805103 7 nips-2005-A Cortically-Plausible Inverse Problem Solving Method Applied to Recognizing Static and Kinematic 3D Objects
15 0.32385799 67 nips-2005-Extracting Dynamical Structure Embedded in Neural Activity
16 0.31604692 35 nips-2005-Bayesian model learning in human visual perception
17 0.31552643 156 nips-2005-Prediction and Change Detection
18 0.31485155 173 nips-2005-Sensory Adaptation within a Bayesian Framework for Perception
19 0.31181514 51 nips-2005-Correcting sample selection bias in maximum entropy density estimation
20 0.2953651 143 nips-2005-Off-Road Obstacle Avoidance through End-to-End Learning
topicId topicWeight
[(3, 0.029), (10, 0.074), (11, 0.011), (25, 0.02), (27, 0.043), (31, 0.037), (34, 0.054), (39, 0.048), (41, 0.013), (55, 0.027), (60, 0.01), (62, 0.316), (69, 0.075), (73, 0.047), (77, 0.011), (88, 0.061), (91, 0.026)]
simIndex simValue paperId paperTitle
same-paper 1 0.79867238 194 nips-2005-Top-Down Control of Visual Attention: A Rational Account
Author: Michael Shettel, Shaun Vecera, Michael C. Mozer
Abstract: Theories of visual attention commonly posit that early parallel processes extract conspicuous features such as color contrast and motion from the visual field. These features are then combined into a saliency map, and attention is directed to the most salient regions first. Top-down attentional control is achieved by modulating the contribution of different feature types to the saliency map. A key source of data concerning attentional control comes from behavioral studies in which the effect of recent experience is examined as individuals repeatedly perform a perceptual discrimination task (e.g., “what shape is the odd-colored object?”). The robust finding is that repetition of features of recent trials (e.g., target color) facilitates performance. We view this facilitation as an adaptation to the statistical structure of the environment. We propose a probabilistic model of the environment that is updated after each trial. Under the assumption that attentional control operates so as to make performance more efficient for more likely environmental states, we obtain parsimonious explanations for data from four different experiments. Further, our model provides a rational explanation for why the influence of past experience on attentional control is short lived. 1 INTRODUCTION The brain does not have the computational capacity to fully process the massive quantity of information provided by the eyes. Selective attention operates to filter the spatiotemporal stream to a manageable quantity. Key to understanding the nature of attention is discovering the algorithm governing selection, i.e., understanding what information will be selected and what will be suppressed. Selection is influenced by attributes of the spatiotemporal stream, often referred to as bottom-up contributions to attention. For example, attention is drawn to abrupt onsets, motion, and regions of high contrast in brightness and color. Most theories of attention posit that some visual information processing is performed preattentively and in parallel across the visual field. This processing extracts primitive visual features such as color and motion, which provide the bottom-up cues for attentional guidance. However, attention is not driven willy nilly by these cues. The deployment of attention can be modulated by task instructions, current goals, and domain knowledge, collectively referred to as top-down contributions to attention. How do bottom-up and top-down contributions to attention interact? Most psychologically and neurobiologically motivated models propose a very similar architecture in which information from bottom-up and top-down sources combines in a saliency (or activation) map (e.g., Itti et al., 1998; Koch & Ullman, 1985; Mozer, 1991; Wolfe, 1994). The saliency map indicates, for each location in the visual field, the relative importance of that location. Attention is drawn to the most salient locations first. Figure 1 sketches the basic architecture that incorporates bottom-up and top-down contributions to the saliency map. The visual image is analyzed to extract maps of primitive features such as color and orientation. Associated with each location in a map is a scalar visual image horizontal primitive feature maps vertical green top-down gains red saliency map FIGURE 1. An attentional saliency map constructed from bottom-up and top-down information bottom-up activations FIGURE 2. Sample display from Experiment 1 of Maljkovic and Nakayama (1994) response or activation indicating the presence of a particular feature. Most models assume that responses are stronger at locations with high local feature contrast, consistent with neurophysiological data, e.g., the response of a red feature detector to a red object is stronger if the object is surrounded by green objects. The saliency map is obtained by taking a sum of bottom-up activations from the feature maps. The bottom-up activations are modulated by a top-down gain that specifies the contribution of a particular map to saliency in the current task and environment. Wolfe (1994) describes a heuristic algorithm for determining appropriate gains in a visual search task, where the goal is to detect a target object among distractor objects. Wolfe proposes that maps encoding features that discriminate between target and distractors have higher gains, and to be consistent with the data, he proposes limits on the magnitude of gain modulation and the number of gains that can be modulated. More recently, Wolfe et al. (2003) have been explicit in proposing optimization as a principle for setting gains given the task definition and stimulus environment. One aspect of optimizing attentional control involves configuring the attentional system to perform a given task; for example, in a visual search task for a red vertical target among green vertical and red horizontal distractors, the task definition should result in a higher gain for red and vertical feature maps than for other feature maps. However, there is a more subtle form of gain modulation, which depends on the statistics of display environments. For example, if green vertical distractors predominate, then red is a better discriminative cue than vertical; and if red horizontal distractors predominate, then vertical is a better discriminative cue than red. In this paper, we propose a model that encodes statistics of the environment in order to allow for optimization of attentional control to the structure of the environment. Our model is designed to address a key set of behavioral data, which we describe next. 1.1 Attentional priming phenomena Psychological studies involve a sequence of experimental trials that begin with a stimulus presentation and end with a response from the human participant. Typically, trial order is randomized, and the context preceding a trial is ignored. However, in sequential studies, performance is examined on one trial contingent on the past history of trials. These sequential studies explore how experience influences future performance. Consider a the sequential attentional task of Maljkovic and Nakayama (1994). On each trial, the stimulus display (Figure 2) consists of three notched diamonds, one a singleton in color—either green among red or red among green. The task is to report whether the singleton diamond, referred to as the target, is notched on the left or the right. The task is easy because the singleton pops out, i.e., the time to locate the singleton does not depend on the number of diamonds in the display. Nonetheless, the response time significantly depends on the sequence of trials leading up to the current trial: If the target is the same color on the cur- rent trial as on the previous trial, response time is roughly 100 ms faster than if the target is a different color on the current trial. Considering that response times are on the order of 700 ms, this effect, which we term attentional priming, is gigantic in the scheme of psychological phenomena. 2 ATTENTIONAL CONTROL AS ADAPTATION TO THE STATISTICS OF THE ENVIRONMENT We interpret the phenomenon of attentional priming via a particular perspective on attentional control, which can be summarized in two bullets. • The perceptual system dynamically constructs a probabilistic model of the environment based on its past experience. • Control parameters of the attentional system are tuned so as to optimize performance under the current environmental model. The primary focus of this paper is the environmental model, but we first discuss the nature of performance optimization. The role of attention is to make processing of some stimuli more efficient, and consequently, the processing of other stimuli less efficient. For example, if the gain on the red feature map is turned up, processing will be efficient for red items, but competition from red items will reduce the efficiency for green items. Thus, optimal control should tune the system for the most likely states of the world by minimizing an objective function such as: J(g) = ∑ P ( e )RT g ( e ) (1) e where g is a vector of top-down gains, e is an index over environmental states, P(.) is the probability of an environmental state, and RTg(.) is the expected response time—assuming a constant error rate—to the environmental state under gains g. Determining the optimal gains is a challenge because every gain setting will result in facilitation of responses to some environmental states but hindrance of responses to other states. The optimal control problem could be solved via direct reinforcement learning, but the rapidity of human learning makes this possibility unlikely: In a variety of experimental tasks, evidence suggests that adaptation to a new task or environment can occur in just one or two trials (e.g., Rogers & Monsell, 1996). Model-based reinforcement learning is an attractive alternative, because given a model, optimization can occur without further experience in the real world. Although the number of real-world trials necessary to achieve a given level of performance is comparable for direct and model-based reinforcement learning in stationary environments (Kearns & Singh, 1999), naturalistic environments can be viewed as highly nonstationary. In such a situation, the framework we suggest is well motivated: After each experience, the environment model is updated. The updated environmental model is then used to retune the attentional system. In this paper, we propose a particular model of the environment suitable for visual search tasks. Rather than explicitly modeling the optimization of attentional control by setting gains, we assume that the optimization process will serve to minimize Equation 1. Because any gain adjustment will facilitate performance in some environmental states and hinder performance in others, an optimized control system should obtain faster reaction times for more probable environmental states. This assumption allows us to explain experimental results in a minimal, parsimonious framework. 3 MODELING THE ENVIRONMENT Focusing on the domain of visual search, we characterize the environment in terms of a probability distribution over configurations of target and distractor features. We distinguish three classes of features: defining, reported, and irrelevant. To explain these terms, consider the task of searching a display of size varying, colored, notched diamonds (Figure 2), with the task of detecting the singleton in color and judging the notch location. Color is the defining feature, notch location is the reported feature, and size is an irrelevant feature. To simplify the exposition, we treat all features as having discrete values, an assumption which is true of the experimental tasks we model. We begin by considering displays containing a single target and a single distractor, and shortly generalize to multidistractor displays. We use the framework of Bayesian networks to characterize the environment. Each feature of the target and distractor is a discrete random variable, e.g., Tcolor for target color and Dnotch for the location of the notch on the distractor. The Bayes net encodes the probability distribution over environmental states; in our working example, this distribution is P(Tcolor, Tsize, Tnotch, Dcolor, Dsize, Dnotch). The structure of the Bayes net specifies the relationships among the features. The simplest model one could consider would be to treat the features as independent, illustrated in Figure 3a for singleton-color search task. The opposite extreme would be the full joint distribution, which could be represented by a look up table indexed by the six features, or by the cascading Bayes net architecture in Figure 3b. The architecture we propose, which we’ll refer to as the dominance model (Figure 3c), has an intermediate dependency structure, and expresses the joint distribution as: P(Tcolor)P(Dcolor |Tcolor)P(Tsize |Tcolor)P(Tnotch |Tcolor)P(Dsize |Dcolor)P(Dnotch |Tcolor). The structured model is constructed based on three rules. 1. The defining feature of the target is at the root of the tree. 2. The defining feature of the distractor is conditionally dependent on the defining feature of the target. We refer to this rule as dominance of the target over the distractor. 3. The reported and irrelevant features of target (distractor) are conditionally dependent on the defining feature of the target (distractor). We refer to this rule as dominance of the defining feature over nondefining features. As we will demonstrate, the dominance model produces a parsimonious account of a wide range of experimental data. 3.1 Updating the environment model The model’s parameters are the conditional distributions embodied in the links. In the example of Figure 3c with binary random variables, the model has 11 parameters. However, these parameters are determined by the environment: To be adaptive in nonstationary environments, the model must be updated following each experienced state. We propose a simple exponentially weighted averaging approach. For two variables V and W with observed values v and w on trial t, a conditional distribution, P t ( V = u W = w ) = δ uv , is (a) Tcolor Dcolor Tsize Tnotch (b) Tcolor Dcolor Dsize Tsize Dnotch Tnotch (c) Tcolor Dcolor Dsize Tsize Dsize Dnotch Tnotch Dnotch FIGURE 3. Three models of a visual-search environment with colored, notched, size-varying diamonds. (a) feature-independence model; (b) full-joint model; (c) dominance model. defined, where δ is the Kronecker delta. The distribution representing the environment E following trial t, denoted P t , is then updated as follows: E E P t ( V = u W = w ) = αP t – 1 ( V = u W = w ) + ( 1 – α )P t ( V = u W = w ) (2) for all u, where α is a memory constant. Note that no update is performed for values of W other than w. An analogous update is performed for unconditional distributions. E How the model is initialized—i.e., specifying P 0 —is irrelevant, because all experimental tasks that we model, participants begin the experiment with many dozens of practice trials. E Data is not collected during practice trials. Consequently, any transient effects of P 0 do E not impact the results. In our simulations, we begin with a uniform distribution for P 0 , and include practice trials as in the human studies. Thus far, we’ve assumed a single target and a single distractor. The experiments that we model involve multiple distractors. The simple extension we require to handle multiple distractors is to define a frequentist probability for each distractor feature V, P t ( V = v W = w ) = C vw ⁄ C w , where C vw is the count of co-occurrences of feature values v and w among the distractors, and C w is the count of w. Our model is extremely simple. Given a description of the visual search task and environment, the model has only a single degree of freedom, α . In all simulations, we fix α = 0.75 ; however, the choice of α does not qualitatively impact any result. 4 SIMULATIONS In this section, we show that the model can explain a range of data from four different experiments examining attentional priming. All experiments measure response times of participants. On each trial, the model can be used to obtain a probability of the display configuration (the environmental state) on that trial, given the history of trials to that point. Our critical assumption—as motivated earlier—is that response times monotonically decrease with increasing probability, indicating that visual information processing is better configured for more likely environmental states. The particular relationship we assume is that response times are linear in log probability. This assumption yields long response time tails, as are observed in all human studies. 4.1 Maljkovic and Nakayama (1994, Experiment 5) In this experiment, participants were asked to search for a singleton in color in a display of three red or green diamonds. Each diamond was notched on either the left or right side, and the task was to report the side of the notch on the color singleton. The well-practiced participants made very few errors. Reaction time (RT) was examined as a function of whether the target on a given trial is the same or different color as the target on trial n steps back or ahead. Figure 4 shows the results, with the human RTs in the left panel and the simulation log probabilities in the right panel. The horizontal axis represents n. Both graphs show the same outcome: repetition of target color facilitates performance. This influence lasts only for a half dozen trials, with an exponentially decreasing influence further into the past. In the model, this decreasing influence is due to the exponential decay of recent history (Equation 2). Figure 4 also shows that—as expected—the future has no influence on the current trial. 4.2 Maljkovic and Nakayama (1994, Experiment 8) In the previous experiment, it is impossible to determine whether facilitation is due to repetition of the target’s color or the distractor’s color, because the display contains only two colors, and therefore repetition of target color implies repetition of distractor color. To unconfound these two potential factors, an experiment like the previous one was con- ducted using four distinct colors, allowing one to examine the effect of repeating the target color while varying the distractor color, and vice versa. The sequence of trials was composed of subsequences of up-to-six consecutive trials with either the target or distractor color held constant while the other color was varied trial to trial. Following each subsequence, both target and distractors were changed. Figure 5 shows that for both humans and the simulation, performance improves toward an asymptote as the number of target and distractor repetitions increases; in the model, the asymptote is due to the probability of the repeated color in the environment model approaching 1.0. The performance improvement is greater for target than distractor repetition; in the model, this difference is due to the dominance of the defining feature of the target over the defining feature of the distractor. 4.3 Huang, Holcombe, and Pashler (2004, Experiment 1) Huang et al. (2004) and Hillstrom (2000) conducted studies to determine whether repetitions of one feature facilitate performance independently of repetitions of another feature. In the Huang et al. study, participants searched for a singleton in size in a display consisting of lines that were short and long, slanted left or right, and colored white or black. The reported feature was target slant. Slant, size, and color were uncorrelated. Huang et al. discovered that repeating an irrelevant feature (color or orientation) facilitated performance, but only when the defining feature (size) was repeated. As shown in Figure 6, the model replicates human performance, due to the dominance of the defining feature over the reported and irrelevant features. 4.4 Wolfe, Butcher, Lee, and Hyde (2003, Experiment 1) In an empirical tour-de-force, Wolfe et al. (2003) explored singleton search over a range of environments. The task is to detect the presence or absence of a singleton in displays conHuman data Different Color 600 Different Color 590 580 570 15 13 11 9 7 Past 5 3.2 3 Same Color 2.8 Same Color 560 550 Simulation 3.4 log(P(trial)) Reaction Time (msec) 610 3 1 +1 +3 +5 Future 2.6 +7 15 13 Relative Trial Number 11 9 7 Past 5 3 1 +1 +3 +5 Future +7 Relative Trial Number FIGURE 4. Experiment 5 of Maljkovic and Nakayama (1994): performance on a given trial conditional on the color of the target on a previous or subsequent trial. Human data is from subject KN. 650 6 Distractors Same 630 5.5 log(P(trial)) FIGURE 5. Experiment 8 of Maljkovic and Nakayama (1994). (left panel) human data, average of subjects KN and SS; (right panel) simulation Reaction Time (msec) 640 620 Target Same 610 5 Distractors Same 4.5 4 600 Target Same 3.5 590 3 580 1 2 3 4 5 1 6 4 5 6 4 1000 Size Alternate Size Alternate log(P(trial)) Reaction Time (msec) 3 4.2 1050 FIGURE 6. Experiment 1 of Huang, Holcombe, & Pashler (2004). (left panel) human data; (right panel) simulation 2 Order in Sequence Order in Sequence 950 3.8 3.6 3.4 900 3.2 Size Repeat 850 Size Repeat 3 Color Repeat Color Alternate Color Repeat Color Alternate sisting of colored (red or green), oriented (horizontal or vertical) lines. Target-absent trials were used primarily to ensure participants were searching the display. The experiment examined seven experimental conditions, which varied in the amount of uncertainty as to the target identity. The essential conditions, from least to most uncertainty, are: blocked (e.g., target always red vertical among green horizontals), mixed feature (e.g., target always a color singleton), mixed dimension (e.g., target either red or vertical), and fully mixed (target could be red, green, vertical, or horizontal). With this design, one can ascertain how uncertainty in the environment and in the target definition influence task difficulty. Because the defining feature in this experiment could be either color or orientation, we modeled the environment with two Bayes nets—one color dominant and one orientation dominant—and performed model averaging. A comparison of Figures 7a and 7b show a correspondence between human RTs and model predictions. Less uncertainty in the environment leads to more efficient performance. One interesting result from the model is its prediction that the mixed-feature condition is easier than the fully-mixed condition; that is, search is more efficient when the dimension (i.e., color vs. orientation) of the singleton is known, even though the model has no abstract representation of feature dimensions, only feature values. 4.5 Optimal adaptation constant In all simulations so far, we fixed the memory constant. From the human data, it is clear that memory for recent experience is relatively short lived, on the order of a half dozen trials (e.g., left panel of Figure 4). In this section we provide a rational argument for the short duration of memory in attentional control. Figure 7c shows mean negative log probability in each condition of the Wolfe et al. (2003) experiment, as a function of α . To assess these probabilities, for each experimental condition, the model was initialized so that all of the conditional distributions were uniform, and then a block of trials was run. Log probability for all trials in the block was averaged. The negative log probability (y axis of the Figure) is a measure of the model’s misprediction of the next trial in the sequence. For complex environments, such as the fully-mixed condition, a small memory constant is detrimental: With rapid memory decay, the effective history of trials is a high-variance sample of the distribution of environmental states. For simple environments, a large memory constant is detrimental: With slow memory decay, the model does not transition quickly from the initial environmental model to one that reflects the statistics of a new environment. Thus, the memory constant is constrained by being large enough that the environment model can hold on to sufficient history to represent complex environments, and by being small enough that the model adapts quickly to novel environments. If the conditions in Wolfe et al. give some indication of the range of naturalistic environments an agent encounters, we have a rational account of why attentional priming is so short lived. Whether priming lasts 2 trials or 20, the surprising empirical result is that it does not last 200 or 2000 trials. Our rational argument provides a rough insight into this finding. (a) fully mixed mixed feature mixed dimension blocked 460 (c) Simulation fully mixed mixed feature mixed dimension blocked 4 5 420 log(P(trial)) 440 2 Blocked Red or Vertical Blocked Red and Vertical Mixed Feature Mixed Dimension Fully Mixed 4 3 log(P(trial)) reaction time (msec) (b) Human Data 480 3 2 1 400 1 380 0 360 0 red or vert red and vert target type red or vert red and vert target type 0 0.5 0.8 0.9 0.95 0.98 Memory Constant FIGURE 7. (a) Human data for Wolfe et al. (2003), Experiment 1; (b) simulation; (c) misprediction of model (i.e., lower y value = better) as a function of α for five experimental condition 5 DISCUSSION The psychological literature contains two opposing accounts of attentional priming and its relation to attentional control. Huang et al. (2004) and Hillstrom (2000) propose an episodic account in which a distinct memory trace—representing the complete configuration of features in the display—is laid down for each trial, and priming depends on configural similarity of the current trial to previous trials. Alternatively, Maljkovic and Nakayama (1994) and Wolfe et al. (2003) propose a feature-strengthening account in which detection of a feature on one trial increases its ability to attract attention on subsequent trials, and priming is proportional to the number of overlapping features from one trial to the next. The episodic account corresponds roughly to the full joint model (Figure 3b), and the feature-strengthening account corresponds roughly to the independence model (Figure 3a). Neither account is adequate to explain the range of data we presented. However, an intermediate account, the dominance model (Figure 3c), is not only sufficient, but it offers a parsimonious, rational explanation. Beyond the model’s basic assumptions, it has only one free parameter, and can explain results from diverse experimental paradigms. The model makes a further theoretical contribution. Wolfe et al. distinguish the environments in their experiment in terms of the amount of top-down control available, implying that different mechanisms might be operating in different environments. However, in our account, top-down control is not some substance distributed in different amounts depending on the nature of the environment. Our account treats all environments uniformly, relying on attentional control to adapt to the environment at hand. We conclude with two limitations of the present work. First, our account presumes a particular network architecture, instead of a more elegant Bayesian approach that specifies priors over architectures, and performs automatic model selection via the sequence of trials. We did explore such a Bayesian approach, but it was unable to explain the data. Second, at least one finding in the literature is problematic for the model. Hillstrom (2000) occasionally finds that RTs slow when an irrelevant target feature is repeated but the defining target feature is not. However, because this effect is observed only in some experiments, it is likely that any model would require elaboration to explain the variability. ACKNOWLEDGEMENTS We thank Jeremy Wolfe for providing the raw data from his experiment for reanalysis. This research was funded by NSF BCS Award 0339103. REFERENCES Huang, L, Holcombe, A. O., & Pashler, H. (2004). Repetition priming in visual search: Episodic retrieval, not feature priming. Memory & Cognition, 32, 12–20. Hillstrom, A. P. (2000). Repetition effects in visual search. Perception & Psychophysics, 62, 800-817. Itti, L., Koch, C., & Niebur, E. (1998). A model of saliency-based visual attention for rapid scene analysis. IEEE Trans. Pattern Analysis & Machine Intelligence, 20, 1254–1259. Kearns, M., & Singh, S. (1999). Finite-sample convergence rates for Q-learning and indirect algorithms. In Advances in Neural Information Processing Systems 11 (pp. 996–1002). Cambridge, MA: MIT Press. Koch, C. and Ullman, S. (1985). Shifts in selective visual attention: towards the underlying neural circuitry. Human Neurobiology, 4, 219–227. Maljkovic, V., & Nakayama, K. (1994). Priming of pop-out: I. Role of features. Mem. & Cognition, 22, 657-672. Mozer, M. C. (1991). The perception of multiple objects: A connectionist approach. Cambridge, MA: MIT Press. Rogers, R. D., & Monsell, S. (1995). The cost of a predictable switch between simple cognitive tasks. Journal of Experimental Psychology: General, 124, 207–231. Wolfe, J.M. (1994). Guided Search 2.0: A Revised Model of Visual Search. Psych. Bull. & Rev., 1, 202–238. Wolfe, J. S., Butcher, S. J., Lee, C., & Hyde, M. (2003). Changing your mind: on the contributions of top-down and bottom-up guidance in visual search for feature singletons. Journal of Exptl. Psychology: Human Perception & Performance, 29, 483-502.
2 0.6177603 60 nips-2005-Dynamic Social Network Analysis using Latent Space Models
Author: Purnamrita Sarkar, Andrew W. Moore
Abstract: This paper explores two aspects of social network modeling. First, we generalize a successful static model of relationships into a dynamic model that accounts for friendships drifting over time. Second, we show how to make it tractable to learn such models from data, even as the number of entities n gets large. The generalized model associates each entity with a point in p-dimensional Euclidian latent space. The points can move as time progresses but large moves in latent space are improbable. Observed links between entities are more likely if the entities are close in latent space. We show how to make such a model tractable (subquadratic in the number of entities) by the use of appropriate kernel functions for similarity in latent space; the use of low dimensional kd-trees; a new efficient dynamic adaptation of multidimensional scaling for a first pass of approximate projection of entities into latent space; and an efficient conjugate gradient update rule for non-linear local optimization in which amortized time per entity during an update is O(log n). We use both synthetic and real-world data on upto 11,000 entities which indicate linear scaling in computation time and improved performance over four alternative approaches. We also illustrate the system operating on twelve years of NIPS co-publication data. We present a detailed version of this work in [1]. 1
3 0.41404712 169 nips-2005-Saliency Based on Information Maximization
Author: Neil Bruce, John Tsotsos
Abstract: A model of bottom-up overt attention is proposed based on the principle of maximizing information sampled from a scene. The proposed operation is based on Shannon's self-information measure and is achieved in a neural circuit, which is demonstrated as having close ties with the circuitry existent in the primate visual cortex. It is further shown that the proposed saliency measure may be extended to address issues that currently elude explanation in the domain of saliency based models. Resu lts on natural images are compared with experimental eye tracking data revealing the efficacy of the model in predicting the deployment of overt attention as compared with existing efforts.
4 0.406517 110 nips-2005-Learning Depth from Single Monocular Images
Author: Ashutosh Saxena, Sung H. Chung, Andrew Y. Ng
Abstract: We consider the task of depth estimation from a single monocular image. We take a supervised learning approach to this problem, in which we begin by collecting a training set of monocular images (of unstructured outdoor environments which include forests, trees, buildings, etc.) and their corresponding ground-truth depthmaps. Then, we apply supervised learning to predict the depthmap as a function of the image. Depth estimation is a challenging problem, since local features alone are insufficient to estimate depth at a point, and one needs to consider the global context of the image. Our model uses a discriminatively-trained Markov Random Field (MRF) that incorporates multiscale local- and global-image features, and models both depths at individual points as well as the relation between depths at different points. We show that, even on unstructured scenes, our algorithm is frequently able to recover fairly accurate depthmaps. 1
5 0.40329215 72 nips-2005-Fast Online Policy Gradient Learning with SMD Gain Vector Adaptation
Author: Jin Yu, Douglas Aberdeen, Nicol N. Schraudolph
Abstract: Reinforcement learning by direct policy gradient estimation is attractive in theory but in practice leads to notoriously ill-behaved optimization problems. We improve its robustness and speed of convergence with stochastic meta-descent, a gain vector adaptation method that employs fast Hessian-vector products. In our experiments the resulting algorithms outperform previously employed online stochastic, offline conjugate, and natural policy gradient methods. 1
6 0.39954633 8 nips-2005-A Criterion for the Convergence of Learning with Spike Timing Dependent Plasticity
7 0.3983978 177 nips-2005-Size Regularized Cut for Data Clustering
8 0.39815977 67 nips-2005-Extracting Dynamical Structure Embedded in Neural Activity
9 0.39531854 144 nips-2005-Off-policy Learning with Options and Recognizers
10 0.39474785 132 nips-2005-Nearest Neighbor Based Feature Selection for Regression and its Application to Neural Activity
11 0.39445776 200 nips-2005-Variable KD-Tree Algorithms for Spatial Pattern Search and Discovery
12 0.39340332 45 nips-2005-Conditional Visual Tracking in Kernel Space
13 0.39335808 170 nips-2005-Scaling Laws in Natural Scenes and the Inference of 3D Shape
14 0.39313936 153 nips-2005-Policy-Gradient Methods for Planning
15 0.39306211 193 nips-2005-The Role of Top-down and Bottom-up Processes in Guiding Eye Movements during Visual Search
16 0.39065269 63 nips-2005-Efficient Unsupervised Learning for Localization and Detection in Object Categories
17 0.38948524 30 nips-2005-Assessing Approximations for Gaussian Process Classification
18 0.3892329 34 nips-2005-Bayesian Surprise Attracts Human Attention
19 0.38692534 173 nips-2005-Sensory Adaptation within a Bayesian Framework for Perception
20 0.38684765 56 nips-2005-Diffusion Maps, Spectral Clustering and Eigenfunctions of Fokker-Planck Operators