Peak-at-end rule: adaptive mechanism predicts time-dependent decision weighting

Abstract

Humans make decisions under various natural circumstances, integrating multiple pieces of information that are distributed over space and time. Although psychophysical and physiological studies have investigated temporal dynamics underlying perceptual decision making, weighting profiles for inliers and outliers during temporal integration have yet to be fully investigated in most studies. Here, we examined the temporal weighting profile of a computational model characterized by a leaky integrator of sensory evidence. As a corollary of its leaky nature, the model predicts the recency effect and overweights outlying elements around the end of the stream. Moreover, we found that the model underweights outlying values occurring earlier in the stream (i.e., robust averaging). We also show that human observers exhibit exactly the same weighting profile in an average estimation task. These findings suggest that the adaptive decision process in the brain results in the time-dependent decision weighting, the “peak-at-end” rule, rather than the peak-end rule in behavioral economics.

Spatial attention in perceptual decision making as revealed by response-locked classification image analysis

“…Eckstein et al 36 applied the CI method to Posner's cueing paradigm 13 and showed that the weight of information in the CI is greater at the spatial location where attention was directed 36 . However, observers in these studies made judgments after the visual stimuli had been shown, like in many psychophysical reverse-correlation studies [28][29][30]32,33 . Such post-stimulus judgment, usually based on the visual working memory, is not necessarily representative of the on-the-fly judgments that we make in real life.…”

“…In visual neuroscience, reverse correlation analysis has been widely applied to reveal information that determines the system response 21,22 . This analysis has been applied not only to the responses of cortical neurons [23][24][25] but also to human behavioral responses for a variety of visual tasks [26][27][28][29][30][31][32][33] . As a variant of the analysis, the classification image (CI) method allows the visualization of what information in the stimuli observers consider important for a given perceptual judgment 34 .…”

Randomised controlled trial of an accommodative support lens designed for computer users

“…In contrast to the main findings, at the end of six months when the SV group wore AS lenses for one week, significantly more participants expressed a preference for AS than for SV lenses. However, this phase only involved the AS group and placebo 56 or recency 57 effects could have confounded the result. The finding in other research 35 of an immediate benefit from a +0.75 D add in CVS in a double‐masked randomised controlled trial warrants further investigation.…”

Spatial attention in perceptual decision making as revealed by response-locked classification image analysis

“…In visual neuroscience, reverse correlation analysis has been widely applied to reveal information that determines the system response [16, 17]. This analysis has been applied not only to the responses of cortical neurons [18, 19, 20] but also to human behavioral responses for a variety of visual tasks [21, 22, 23, 24, 25, 26, 27, 28]. As a variant of the analysis, the classification image (CI) method allows the visualization of what information in the stimuli observers consider important for a given perceptual judgment [29].…”

“…Eckstein et al (2002) applied the CI method to Posner's cueing paradigm [8] and showed that the weight of information in the CI is greater at the spatial location where attention was directed [31]. However, observers made judgments after the visual stimuli had been shown, like in many psychophysical reverse-correlation studies [23,24,25,27,28]. Such post-stimulus judgment, usually based on the visual working memory, is not necessarily representative of the on-the-fly judgments that we make in real life.…”