“Taste typicality” is a foundational and multi-modal dimension of ordinary aesthetic experience

Beauty is in the eye of your cohort: Structured individual differences allow predictions of individualized aesthetic ratings of images

In the last few years, there has been an increasing interest in computational models that are capable of predicting the aesthetic ratings of images based on objective image features. Given that aesthetic ratings vary across individuals, models that predict the average aesthetic ratings ignore the unique taste of an individual. In this paper, our goal is to better understand the individual differences in aesthetic ratings by investigating if individual differences follow structural rules or if taste is due to a random component of an individual's ratings. We address this question by using a collaborative filtering model that uses the similarities in ratings of a cohort of observers to predict individuals' ratings on a new set of images. Using Amazon Mechanical Turk, 299 online participants were instructed to rate how much they like a set of 50 art images. Using a subset of the images (40), we formed cohorts of individuals with similar ratings and used these cohorts to predict how each person would rate the remaining 10 images not included in the training set. The selected cohorts predicted individual ratings significantly better than random cohorts and outperformed predictions based on the mean image ratings. We also found that the optimal size was approximately 12 % of the sample size. These results imply that individual differences in fact have an underlying structure that is consistent across the cohort and are not random. Using personality scores and subject backgrounds, we also looked at the subject characteristics of the cohorts and found that the participants' art background was the only significant factor. Finally, we explored whether the cohorts used particular visual features in a consistent way. For our small set of features, we didn't find any evidence for this. These results provide important insights into the sources of individual differences in aesthetic preferences and highlight the potential for computational models to improve predictions of individual preferences by leveraging structured individual differences rather than relying solely on population averages.

The perceptual primacy of feeling: Affectless visual machines explain a majority of variance in human visually evoked affect

Looking at the world often involves not just seeing things, but feeling things. Modern feedforward machine vision systems that learn to perceive the world in the absence of active physiology, deliberative thought, or any form of feedback that resembles human affective experience offer tools to demystify the relationship between seeing and feeling, and to assess how much of visually evoked affective experiences may be a straightforward function of representation learning over natural image statistics. In this work, we deploy a diverse sample of 180 state-of-the-art deep neural network models trained only on canonical computer vision tasks to predict human ratings of arousal, valence, and beauty for images from multiple categories (objects, faces, landscapes, art) across two datasets. Importantly, we use the features of these models without additional learning, linearly decoding human affective responses from network activity in much the same way neuroscientists decode information from neural recordings. Aggregate analysis across our survey, demonstrates that predictions from purely perceptual models explain a majority of the explainable variance in average ratings of arousal, valence, and beauty alike. Finer-grained analysis within our survey (e.g. comparisons between shallower and deeper layers, or between randomly initialized, category-supervised, and self-supervised models) point to rich, preconceptual abstraction (learned from diversity of visual experience) as a key driver of these predictions. Taken together, these results provide further computational evidence for an information-processing account of visually evoked affect linked directly to efficient representation learning over natural image statistics, and hint at a computational locus of affective and aesthetic valuation immediately proximate to perception.

Consensus and contention in beauty judgment

Variance across participants is at the heart of the centuries-old debate about the universality of beauty. Beauty's belonging to the eye of the beholder implies large interindividual variance, while beauty as a universal object property implies the opposite. To characterize the variance at the center of this debate, we selected two quartets with either high- or low-variance images with high typicality and a given mean beauty. The quartets have high or low variance across 50 participants (group variance) and correspondingly high or low variance across images of a quartet for each participant (quartet variance). We asked 52 new participants to estimate their own mean and quartet variance. Participants successfully predicted their quartet mean but failed to predict their quartet variance. Though invisible, beauty variance is essential to prediction, both in theory and in practice. The quartets show that mean beauty is not the whole story - beauty variance is heterogeneous.

Marketing sonified fragrance: Designing soundscapes for scent

Auditory branding is undoubtedly becoming more important across a range of sectors. One area, in particular, that has recently seen significant growth concerns the introduction of music and soundscapes that have been specifically designed to match a particular scent (what one might think of as "audio scents" or "sonic scents"). This represents an exciting new approach to the sensory marketing of fragrance and for industries with strategic sensory goals, such as cosmetics. Crucially, techniques such as the semantic differential technique, as well as the emerging literature on crossmodal correspondences, offer both a mechanistic understanding of, and a practical framework for, those wishing to rigorously align the connotative meaning and conceptual/emotional/sensory associations of sound and scent. These developments have enabled those working in the creative industries to start moving beyond previously popular approaches to matching, or translating between the senses, that were traditionally often based on the idiosyncratic phenomenon of synaesthesia, toward a more scientific approach while nevertheless still enabling/requiring a healthy dose of artistic inspiration. In this narrative historical review, we highlight the various approaches to the systematic matching of sound with scent and review the various marketing activations that have appeared in this space recently.

Exploring quantitative measures in metacognition of emotion

Metacognition of emotion (meta-emotion) refers to the ability to evaluate and identify one's emotional feelings. No previous study has defined and measured this construct through objective and quantitative procedures. We established a reliable method to measure meta-emotion. With a two-interval forced-choice procedure, participants selected which of two pictures elicited stronger positive emotion; via the Law of Comparative Judgment, their responses were used to compute individual psychological distances for the emotional responses triggered by the pictures. Then, participants were asked to judge whether a pre-exposed picture induced a stronger positive emotion than the median of that elicited by the whole picture set, followed by a confidence rating. By utilizing each individual's psychological distance, the correctness of a participant's emotional experience was quantified by d', and meta-emotion was quantified using meta-d', M-ratio, and M-diff as indices of metacognitive sensitivity and efficiency based on Signal-Detection Theory. Test-retest reliabilities, validated by Spearman correlation, were observed in meta-d', M-ratio, and marginally with M-diff, suggesting the stability of meta-emotion in the current design. This study unveils a validated procedure to quantify meta-emotion, extendable for assessing metacognition of other subjective feelings. Nevertheless, caution is warranted in interpretation, as the measured processes may be influenced by non-metacognitive factors.

Genetic effects on variability in visual aesthetic evaluations are partially shared across visual domains

The aesthetic values that individuals place on visual images are formed and shaped over a lifetime. However, whether the formation of visual aesthetic value is solely influenced by environmental exposure is still a matter of debate. Here, we considered differences in aesthetic value emerging across three visual domains: abstract images, scenes, and faces. We examined variability in two major dimensions of ordinary aesthetic experiences: taste-typicality and evaluation-bias. We build on two samples from the Australian Twin Registry where 1547 and 1231 monozygotic and dizygotic twins originally rated visual images belonging to the three domains. Genetic influences explained 26% to 41% of the variance in taste-typicality and evaluation-bias. Multivariate analyses showed that genetic effects were partially shared across visual domains. Results indicate that the heritability of major dimensions of aesthetic evaluations is comparable to that of other complex social traits, albeit lower than for other complex cognitive traits. The exception was taste-typicality for abstract images, for which we found only shared and unique environmental influences. Our study reveals that diverse sources of genetic and environmental variation influence the formation of aesthetic value across distinct visual domains and provides improved metrics to assess inter-individual differences in aesthetic value.

Aesthetic preferences for prototypical movements in human actions

A commonplace sight is seeing other people walk. Our visual system specializes in processing such actions. Notably, we are not only quick to recognize actions, but also quick to judge how elegantly (or not) people walk. What movements appear appealing, and why do we have such aesthetic experiences? Do aesthetic preferences for body movements arise simply from perceiving others' positive emotions? To answer these questions, we showed observers different point-light walkers who expressed neutral, happy, angry, or sad emotions through their movements and measured the observers' impressions of aesthetic appeal, emotion positivity, and naturalness of these movements. Three experiments were conducted. People showed consensus in aesthetic impressions even after controlling for emotion positivity, finding prototypical walks more aesthetically pleasing than atypical walks. This aesthetic prototype effect could be accounted for by a computational model in which walking actions are treated as a single category (as opposed to multiple emotion categories). The aesthetic impressions were affected both directly by the objective prototypicality of the movements, and indirectly through the mediation of perceived naturalness. These findings extend the boundary of category learning, and hint at possible functions for action aesthetics.

Beauty isn't special: Comparing the information capacity of beauty and other sensory judgments

Information theory (bits) allows comparing beauty judgment to perceptual judgment on the same absolute scale. In one of the most influential articles in psychology, Miller (1956) found that classifying a stimulus into one of eight or more categories of the attribute transmits roughly 2.6 bits of information. That corresponds to 7 ± 2 categories. This number is both remarkably small and highly conserved across attributes and sensory modalities. This appears to be a signature of one-dimensional perceptual judgment. We wondered whether beauty can break this limit. Beauty judgments matter and play a key role in many of our real-life decisions, large and small. Mutual information is how much information about one variable can be obtained from observing another. We measured the mutual information of 50 participants' beauty ratings of everyday images. The mutual information saturated at 2.3 bits. We also replicated the results using different images. The 2.3 bits conveyed by beauty judgment are close to Miller's 2.6 bits of unidimensional perceptual judgment and far less than the 5 to 14 bits of a multidimensional perceptual judgment. By this measure, beauty judgment acts like a perceptual judgment, such as rating pitch, hue, or loudness.

The intrinsic variance of beauty judgment

Recall memory and sequential dependence threaten the independence of successive beauty ratings. Such independence is usually assumed when using repeated measures to estimate the intrinsic variance of a rating. We call "intrinsic" the variance of all possible responses that the participant could give on a trial. Variance arises within and across participants. In attributing the measured variance to sources, the first step is to assess how much is intrinsic. In seven experiments, we measure how much of the variability across beauty ratings can be attributed to recall memory and sequential dependence. With a set size of one, memory is a problem and contributes half the measured variance. However, we showed that for both beauty and ellipticity, with set size of nine or more, recall memory causes a mere 10% increase in the variance of repeated ratings. Moreover, we showed that as long as the stimuli are diverse (i.e., represent different object categories), sequential dependence does not affect the variance of beauty ratings. Lastly, the variance of beauty ratings increases in proportion to the 0.15 power of stimulus set size. We show that the beauty rating of a stimulus in a diverse set is affected by the stimulus set size and not the value of other stimuli. Overall, we conclude that the variance of repeated ratings is a good way to estimate the intrinsic variance of a beauty rating of a stimulus in a diverse set.

Consonance and dissonance perception. A critical review of the historical sources, multidisciplinary findings, and main hypotheses

Revealed more than two millennia ago by Pythagoras, consonance and dissonance (C/D) are foundational concepts in music theory, perception, and aesthetics. The search for the biological, acoustical, and cultural factors that affect C/D perception has resulted in descriptive accounts inspired by arithmetic, musicological, psychoacoustical or neurobiological frameworks without reaching a consensus. Here, we review the key historical sources and modern multidisciplinary findings on C/D and integrate them into three main hypotheses: the vocal similarity hypothesis (VSH), the psychocultural hypothesis (PH), and the sensorimotor hypothesis (SH). By illustrating the hypotheses-related findings, we highlight their major conceptual, methodological, and terminological shortcomings. Trying to provide a unitary framework for C/D understanding, we put together multidisciplinary research on human and animal vocalizations, which converges to suggest that auditory roughness is associated with distress/danger and, therefore, elicits defensive behavioral reactions and neural responses that indicate aversion. We therefore stress the primacy of vocality and roughness as key factors in the explanation of C/D phenomenon, and we explore the (neuro)biological underpinnings of the attraction-aversion mechanisms that are triggered by C/D stimuli. Based on the reviewed evidence, while the aversive nature of dissonance appears as solidly rooted in the multidisciplinary findings, the attractive nature of consonance remains a somewhat speculative claim that needs further investigation. Finally, we outline future directions for empirical research in C/D, especially regarding cross-modal and cross-cultural approaches.