Emotional judgments of scenes are influenced by unintentional averaging

Abstract

Background: The visual system uses ensemble perception to summarize visual input across a variety of domains. This heuristic operates at multiple levels of vision, compressing information as basic as oriented lines or as complex as emotional faces. Given its pervasiveness, the ensemble unsurprisingly can influence how an individual item is perceived, and vice versa. Methods: In the current experiments, we tested whether the perceived emotional valence of a single scene could be influenced by surrounding, simultaneously presented scenes. Observers first rated the emotional valence of a series of individual scenes. They then saw ensembles of the original images, presented in sets of four, and were cued to rate, for a second time, one of four. Results: Results confirmed that the perceived emotional valence of the cued image was pulled toward the mean emotion of the surrounding ensemble on the majority of trials, even though the ensemble was task-irrelevant. Control experiments and analyses confirmed that the pull was driven by high-level, ensemble information. Conclusion: We conclude that high-level ensemble information can influence how we perceive individual items in a crowd, even when working memory demands are low and the ensemble information is not directly task-relevant.

The pleasure of multiple images

“…Our results thus differ from the ones reported by Alwis and Haberman (2020) who found that people are biased toward the mean when reporting the pleasure of one image among four. This difference is most likely due to differences between our and their experimental paradigm.…”

“…The pleasure of images that were otherwise highly pleasurable (pleasure >4.6, on average across participants) was diminished. This effect does not represent a bias towards the mean-the effect that has been observed when people report other properties of one target among several distractors (e.g., Alwis & Haberman, 2020;Brady & Alvarez, 2011; Whitney, 2009;Maule et al, 2014;Zhang & Luck, 2008). Importantly, we did consider the average-biased model, but found that its performance is consistently worse than the one of either a faithful or a high-pleasure attenuation model (see Fig.…”

“…One recent study found that people's pleasure rating of one scene among three others is biased toward the mean pleasure of the presented scenes (Alwis & Haberman, 2020). That study differs from ours in using a long image presentation (1.5 s), time enough for more than three glimpses, whereas we allowed time (0.2 s) for only a single glimpse.…”

Your ears don’t change what your eyes like: People can independently report the pleasure of music and images.

“…Additionally, prior work found that observers cannot faithfully report the single pleasure of a visual stimulus in an array of four stimuli when presented very briefly (200 ms), but that they instead report overall diminished pleasure (Brielmann & Pelli, 2021). This result stands in contrast to similar work that presented multiple images for a longer duration (1.5 s) and found that observers are biased toward the mean pleasure of the entire array (Alwis & Haberman, 2020). When presented with multiple images for a longer period, observers likely had time to attend to each stimulus in a serial manner, which then subsequently resulted in averaging across all stimuli when making a judgment.…”

“…The authors found that participants could not faithfully report the pleasure of a single image unless cued to which image they would be rating ahead of time, and that observers also did not faithfully report the average pleasure of four images (Brielmann & Pelli, 2021). With longer presentation durations of 1.5 s (vs. 200 ms in Brielmann & Pelli, 2020) and a homogeneous image set, pleasure ratings of one scene among three others were found to be biased toward the mean pleasure of the presented scenes (Alwis & Haberman, 2020).…”