Research Statement and General Research Interests

Bryan R. Burnham, Ph.D.

(Current as of September 2009)

My research focuses on attention and information processing. My goal is to further scientific understanding of how humans selectively attend to specific items and locations,  divide attention across and within objects, and the mechanisms humans use to control  attention. Recently, I have become interested in the relationship between political attitudes, androgen exposure revealed through body morphology, and information-processing. Below, I summarize several areas of research that I continually engage in to address these issues.

1. ATTENTIONAL CAPTURE AND THE CONTROL OF SELECTIVE ATTENTION

With the abundance of stimuli in the environment to process at any moment, humans must selectively attend to relevant stimuli and withhold attention from irrelevant stimuli. For example,  the driver of a car should direct attention toward the road and traffic signals  while ignoring any irrelevant and potentially distracting events, such as a clown riding a unicycle on the side of the road. To study this, I have used attentional capture paradigms, in which the presence of simple salient, but irrelevant, visual stimuli (i.e., singletons--an item that differs from its homogeneous background) can influence responding to a target. Specifically, the appearance of a single red item in the same location as a target can speed responding to that target, compared to when the red item and the target are spatially distant. My interest is not the attentional capture effect (observed difference in response times) per se, but rather in the mechanisms by which attentional capture occurs and also the mechanisms that individuals use to selectively ignore salient stimuli.

1A. Attentional Capture by Static Discontinuities: In a 2008 paper published by myself and Jim Neely (Burnham & Neely, 2008), we showed that a static discontinuity that was instantiated at the boundary between to different perceptual groups (a line of red X's juxtaposed next to a line of green O's) captured attention, even though participants were set to identify a dynamic, transient discontinuity (the abrupt onset of a single item). Some theories suggest that attention can be controlled and a person can suppress shifting attention toward an irrelevant discontinuity, and studies have found attentional capture when a target and a singleton are within the same dimension (intra-dimension capture); but few studies have found attentional capture when a singleton and target are defined by different perceptual dimensions (cross-dimension capture). We observed cross-dimension attentional capture by this color/shape discontinuity in five separate experiments. This result appears to be due to the discontinuity creating a boundary between perceptual groups, which received high attentional priority. This research suggests that when humans parse the visual field, boundaries attract attention, because they are important for differentiating between groups of stimuli.

In a paper due to be published in 2009, we (Burnham, Neely, Naginsky & Thomas) demonstrated that the attentional capture alluded to above  was purely stimulus-driven. One issue of contention in the literature is whether a salient singleton automatically captures attention, or if its salient feature needs to be part of an attentional set to capture attention. The results alluded to in the previous paragraph were not automatic because the “singleton” feature of the single abrupt-onset target could have been part of a person’s attentional control settings. Because the static discontinuity was a singleton as well, it may have only captured attention because its “singleton” feature was in the attentional settings. To eliminate the "singletonness" of the target we presented several stimuli simultaneously with the target. If the static discontinuity captured attention, it would be a purely automatic effect, because it was featurally irrelevant to a person’s attentional set, which is what we found in seven separate experiments.

1B. Stimulus-Driven Attentional Capture is Mediated by Working Memory: I have very recently been studying the relationship between working memory and the control of covert orienting of attention. Working memory is a multi-component model that processes information in several different sub-systems that are specific to environmental stimuli that are controlled by a central executive. Because working memory controls information-processing, working memory may be involved with the control of attention. My research has employed visual search tasks where participants locate a target while attempting to ignore a singleton. The critical manipulations have been the load that is placed on working memory during this visual search task. If working memory controls attention, then attentional capture should increase in magnitude when working memory load is high, compared to low. Indeed, in a series of studies in my doctoral dissertation, when the phonological loop component of working memory was loaded (system responsible for processing acoustic information), I found that the attentional capture effect increased in magnitude.

More recently, I have examined the relationship between individual differences in generalized working memory capacity (i.e., the amount of information an individual can process) and attentional control. In this study, participants located a target with a specific visual feature (abrupt onset target for half the participants and red for the other half) that appeared after a singleton cue that was featurally relevant to the target (onset cue for the onset target and a red cue for the red target) or featurally irrelevant to the target (a red cue for the onset target and an onset cue for the red target). Participants also completed the operation span test (OSPAN) of working memory, where larger scores indicate stronger generalized working memory capacity. I found no relationship between working memory (OSPAN) and attentional capture effects elicited by relevant cues; however, individuals with less memory capacity (small OSPAN scores) showed larger attentional capture effects elicited by irrelevant cues compared to individuals with more working memory capacity (high OSPAN scores). The results suggest that working memory acts as an inhibitor that precludes shifting attention toward irrelevant stimuli, not as a facilitator that allows shifting of attention toward relevant stimuli.

1C. Visual Field Asymmetries in Attending to Salient Stimuli: Recently, my lab began examining a difference in processing salient visual singletons between the left and right visual fields. Following up on a poster presented at the 2008 meeting of the Psychonomic Society (Du & Abrams, 2008), we found that when observers identified a target that appeared at a central location, a salient singleton appearing in the left visual field produced a larger attentional capture effect compared to when that  same singleton appeared in the right visual field. Importantly, this asymmetry was observed only when participants were in a very focused state of attention and the target was difficult to discriminate from nontargets. We are currently examining this phenomenon to understand the relationship between top-down and bottom-up control of attention between the two hemispheres.

2. ATTENTION TO OBJECT COMPONENTS

Research has shown that attending to part of an object facilitates attending to other parts of that object, compared to parts on other objects (Egly, Driver & Rafal, 1994). Research has also shown that when dividing attention across objects, perception is better to parts belonging to one object than to several different objects (Duncan, 1984; Law & Abrams, 2002). My research has focused on where attention is distributed within and across objects, and on whether people better are at detecting features on one object, or two different objects.

2.A. Priority is Assigned to Line Intersections (Junctures): Research has found that attention to an object is not distributed equally throughout an object; rather, attention is directed toward an object's corners. Cole, Gellatly and Blurton (2001) demonstrated that the time to detect a probe was faster when it appeared near an object’s corner than along an edge. Although Cole et al. found that corners are attended to more than edges, it might not be the corners on an object per se that are prioritized; rather, it may be that intersections (junctures) capture attention and are prioritized whether they are part of an enclosed object or not. To address this, we had people respond to a target among nontargets that appeared on the circumference of an imaginary circle. Prior to this target display, two intersecting lines that were not part of an enclosed object were presented. The target appeared where the line intersection appeared, at the point where one of the lines was presented but did not intersect with the other line, or away from both lines. People were faster to respond to targets presented near the intersection compared to along a straight edge of one line and away from both lines. This suggests attention was captured by the line intersection and that being part of an enclosed object may not be necessary for intersections to be prioritized. Interestingly, we also found that people were faster when the target appeared where one lines appeared, compared to away from both lines. Initially, we thought this was due to observers creating an “imaginary” intersection between that one line and the circular target display (i.e., an “imaginary” intersection captured attention). However, after additional research, although an “imaginary” intersection did capture some attention, we found that simply being near a line facilitates responding to a target. We are currently examining this "imaginary" intersection effect.

2B. Two Objects are Easier to Attend then One: When dividing attention across objects it should be easier to attend to parts on one object than on different objects. However, G. Davis and colleagues (Davis & Holmes, 2005) found the opposite. In their studies, people were slower to determine whether two features matched when the features appeared on different ends of the same object, compared to on two different objects. However, the Davis and Holmes design was confounded, because the features were twice as likely to appear on different objects as on one object. Thus, the faster responses to features on different objects may have been because people were biased to shift their attention between objects. In our research, we replicated this different object benefit using their stimuli, but without the possible bias. We also obtained this effect when we placed the two features within the bodies of the objects, rather than within the outline of the objects, and when the objects (and features) were presented briefly as well as for longer durations. In short, we manipulated several design characteristics to examine boundary conditions of the different object benefit, which we found to be quite robust. We have conducted a number of studies on this phenomenon and it appears that the effect is due to the features being in the outline of the objects.

SUMMARY

I have conducted a number of experiments on two broad areas of human attention: (1) selective attention to visual stimuli, (2) attention to objects and object components, and (3) political attitudes and information-processing. In the future, I plan to address questions related to the projects listed in my research statement, so as to facilitate our understanding the mechanisms that control the allocation of human visual attention to objects, locations and stimuli in the visual field.