A brain-derived metric for preferred kinetic stimuli

Semir Zeki,Jonathan Stutters

doi:10.1098/rsob.120001

Abstract

We here address the question of whether there is any correlation between subjective preference for simple configurations within a specific visual domain such as motion and strength of activity in visual areas in which that domain is emphasized. We prepared several distinctive patterns of dots in motion with various characteristics and asked humans to rate them according to their preference, before and while scanning the activity in their brains with functional magnetic resonance imaging. For simplicity, we restricted ourselves to motion in the fronto-parallel plane. Moving patterns produced activity in areas V1, V2, the V3 complex (V3, V3A, V3B) and V5, but only in areas V5, V3A/B and parietal cortex did the preferred kinetic patterns produce stronger activity when compared with the non-preferred ones. In addition, preferred patterns produced activity within field A1 of medial orbito-frontal cortex (mOFC), which is not otherwise activated by kinetic stimuli. Hence, for these areas, stronger neural activity correlated with subjective preference. We conclude that configurations of kinetic stimuli that are subjectively preferred correlate with stronger activity within early visual areas and within mOFC. This opens up the possibility of more detailed studies to relate subjective preferences to strength of activity in early visual areas and to relate activity in them to areas whose activity correlates with the subjective experience of beauty.

Highlights

In the work reported here, we address within a specific and restricted context a more general question of whether there are any definable characteristics of stimuli that render them more attractive, or at any rate preferable
Basing ourselves on the functional specialization of the visual brain for different visual attributes [2,3,4], among which is a specialization for visual motion [5,6,7], we asked whether there are any particular patterns of dots in motion that stimulate visual areas known to contain directionally selective cells preferentially
Any visual or cortical area in which strength of activity correlated with strength of subjective preference would have been of interest, we were especially interested in areas containing directionally selective cells or ones that, in the human, respond strongly to visual motion stimulation

Summary

Introduction

The question has been theoretically addressed many times before in artistic speculation, though within a much broader context Characteristics such as harmony, proportion and symmetry have at various times been considered to be attributes of beautiful works, but without a general consensus. What constitutes proportion or symmetry in one category of visual stimuli (e.g. objects) cannot be translated to other attributes (e.g. colour or motion) One way around this difficulty is to concentrate on a single visual attribute, such as visual motion, and enquire whether there are any characteristics or configurations that, for human observers, make some kinetic patterns preferable to others and, if so, whether we can account for this preference in neural terms. The V5 complex (MTþ) is the most prominent, other

Methods

Results

Discussion

Conclusion