Fear Helps You See Better

The study of the influence of emotion on cognition and perception has really taken off over the last decade or so, which is a good thing, because cognitive psychologists pretty much ignored emotion for a long time, so we have some catching up to do. Over at PsyBlog, Jeremy Dean summarizes an Annual Review paper that reviews much of this research. In his summary, Jeremy mentions a paper on emotion and vision that I'd meant to post on a while ago because it's really cool, but forgot all about. So I'll post about it now.

The paper, by Phelps, Ling, and Carrasco1, describes two studies that focus on the influence of fear on perception. There's a great deal of evidence that fear can focus attention, making it easier for us to notice things that we otherwise would have missed. What's not clear, however, is how fear does this. Phelps et al. hypothesize that fear may influence early visual processing, and test this hypothesis by looking at how fear influences contrast sensitivity. Contrast sensitivity refers to our ability to distinguish a disguised signal from noise, and involves processes that occur very early in visual perception. In most experiments on contrast sensitivity, researchers use Gabor patches like this one:

These patches are created using a statistical distribution (that one looks Gaussian), and can be manipulated so that the patch is more or less distinguishable from the gray background. The greater your contrast sensitivity, the less distinguishable the patch has to be from the background for you to pick it out.

In their first experiment, Phelps et al. had participants focus on a fixation point in the middle of a computer screen, and then briefly (for 75 ms) flashed either a fearful or neutral face at the center of the screen (the two types of faces can be seen below, from Phelps et al. Figure 1, p. 294). Viewing fearful faces has been shown to cause a fearful reaction, even when those faces are presented for a very short period of time. After the face was gone, there was a 50 ms delay, at which point four Gabor patches were flashed on the screen for 45 ms. Three of the patches were vertical (distractors) and the fourth was titled 8° from the vertical either to the right or the left (the target stimulus). Across trials, the contrast levels of the target patches were varied. The participants' task was to report whether the target stimulus was tilted to the right or the left.

At this point you may be wondering why the stimuli were presented for such a short period of time (the total time from the first appearance of the face to the end of the presentation of the Gabor patches, including the interval in between the faces and the patches, was less than 2 tenths of a second!). This was to make sure that people were unable to make any volunteer eye movements. This means that if fear influences contrast sensitivity, it will be independent of the influence of fear on willfully allocated attention. The following graph (taken from Figure 2, p. 295) presents the results of Experiment 1:

As you can see, on trials when a fearful face was presented, accuracy was greater across most levels of contrast. The most important thing, though, is that fearful faces were able to determine the orientation of the target patches more accurately at low levels of contrast, indicating that the fearful reaction induced by viewing the fearful faces increased participants' contrast sensitivity for subsequently presented patches.

In Experiment 2, Phelps et al. were able to show that this effect is due to an interaction between fear and "covert" (i.e., reflexive) attention. They did this by either directing participants' attention to a particular spot on the screen, and then briefly presenting the target stimulus there, or presenting the stimulus in a spot on the screen without first directing the participants' attention. If fear is influencing vision independent of reflexive attention, then it shouldn't matter whether the participants' attention has been directed to the spot of the stimulus or not. Either way, fear should improve contrast sensitivity (attention itself also improves contrast sensitivity, so if the two are independent, the effects of fear and attention should be additive). However, if fear interacts with reflexive attention so that attentional effects are heightened, contrast sensitivity should improve significantly when a fearful face is presented only when attention has been directed. This figure presents the results in the form of contrast thresholds, represented by the corresponding Gabor patches (and also gives you an idea of how the Gabor patches can be altered to decrease or increase contrast; from Phelps et al. Figure 4, p. 297):

As you can see, the contrast sensitivity for directed attention (the two patches on the left, labeled "peripheral") is greater than for undirected attention, and when attention is directed, fear further increases contrast sensitivity. When attention is not directed, however, fear has no significant effect on contrast sensitivity.

So, the results of these two experiments indicate that fear can influence early visual processing, and that it does so by improving reflexive attention (it has little or no effect on willfully-guided attention). In short, fear improves our vision, by automatically focusing our attentional resources. While this is very cool, it's not entirely surprising. Adaptively, it's a pretty good idea to be more attentive to fear-inducing stimuli, so it's a good idea to be more attentive when you're afraid. It is a bit surprising that the effects of fear can occur so early in visual processing, but as cognitive scientists are continually learning, it's never a good idea to underestimate the power of emotion at any level of processing.

¹Phelps, E. A., Ling, S., & Carrasco, M. (2006). Emotion facilitates perception and potentiates the perceptual benefit of attention. Psychological Science, 17, 292-299.