A highly selective response to food in human visual cortex revealed by hypothesis-free voxel decomposition

Khosla, Meenakshi; Murty, N. Apurva Ratan; Kanwisher, Nancy

doi:10.1016/j.cub.2022.08.009

Cited by 48 publications

(86 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Together, these results suggest that the positivity and sparsity constraints applied by NMF enable it to recover more robust and interpretable components from human behavioral data than PCA. These benefits are likely to extend to neural data, as suggested by the recent application of NMF to reveal novel category selectivity in human fMRI data (Khosla et al, 2022).…”

Section: Discussionmentioning

confidence: 92%

A data-driven investigation of human action representations

Dima

Hebart

Işık

2022

Preprint

View full text Add to dashboard Cite

Understanding actions performed by others requires us to integrate different types of information about people, scenes, objects, and their interactions. What organizing dimensions does the mind use to make sense of this complex action space? To address this question, we collected intuitive similarity judgments across two large-scale sets of naturalistic videos depicting everyday actions. We used cross-validated sparse non-negative matrix factorization (NMF) to identify the structure underlying action similarity judgments. A low-dimensional representation, consisting of nine to ten dimensions, was sufficient to accurately reconstruct human similarity judgments. The dimensions were robust to stimulus set perturbations and reproducible in a separate odd-one-out experiment. Human labels mapped these dimensions onto semantic axes relating to food, work, and home life; social axes relating to people and emotions; and one visual axis related to scene setting. While highly interpretable, these dimensions did not share a clear one-to-one correspondence with prior hypotheses of action-relevant dimensions. Together, our results reveal a low-dimensional set of robust and interpretable dimensions that organize intuitive action similarity judgments and highlight the importance of data-driven investigations of behavioral representations.

show abstract

Section: Discussionmentioning

confidence: 92%

A data-driven investigation of human action representations

Dima

Hebart

Işık

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Food pictures therefore have been chosen as a proxy to investigate cognitive and brain mechanisms underlying food perception (Killgore et al, 2003;Hare et al, 2009;Simmons et al, 2016). In the past years, several studies have unveiled a network of brain regions active while viewing foods compared to other objects in both human and non-human primates (e.g., Frank et al, 2010;Garcia-Garcia et al, 2013;Khosla et al, 2022;Killgore et al, 2003;Kriegeskorte et al, 2008;Simmons et al, 2016;Tsourides et al, 2016). For instance, using magnetoencephalography (MEG), Tsourides et al (2016) showed how the human brain discriminates between foods (e.g., fruits and vegetables, meat) and inedible stimuli (e.g., rotten foods, artifact objects) as early as 85 ms post-stimulus onset.…”

Section: What To Look Formentioning

confidence: 99%

“…Another recent study conducted by Adamson and Troiani (2018) compared brain responses to foods and faces, another very biologically salient type of stimulus, and found overlapping activity in the visual cortex (in particular in the fusiform gyrus; activity in this region appears to be influenced by the rewarding value of stimuli). In general, brain responses to food images compared to non-food images showed selective activations in the ventral visual cortex, in particular in regions crucial for object recognition (Khosla et al, 2022; Van der Laan et al, 2011 for a meta-analysis). This converging evidence suggests a 'pop-out' effect of food stimuli, which appears to capture our attention, and our cognitive resources are then employed to evaluate such stimuli (Van der Laan et al, 2011).…”

Section: What To Look Formentioning

confidence: 99%

“…Human beings spent most of their history as hunter-gatherers, having to navigate their environment to Find and Evaluate beneficial foods while Excluding the costs of consuming something harmful to finally Decide which entities to consume (see Figure 1). We have called this the FEED problem and we have argued that a focus on FEED can bridge the gap between different findings in the field of food cognition and shed a new light on the cognitive and brain mechanisms that drive our food behaviors In sum, the sections Find and Evaluate reveal that humans are extremely skilled at detecting potential food resources among other entities, with a highly selective and rapid response to food in the ventral visual cortex (around 100 ms after stimulus presentation in experimental setups, e.g., Khosla et al, 2022;Tsourides et al, 2016). Any food candidates in the environment capture human attention (a pop-out effect similar to the one existing for faces, Van der Laan et al, 2011), while certain intrinsic characteristics of foods such as color, caloric content or level of processing modulate perceptual and higher-order cognitive mechanisms (e.g., Foroni & Rumiati, 2017, Hesse & Knight, 2022.…”

Section: Summary and Practical Implicationsmentioning

confidence: 99%

See 1 more Smart Citation

FEED your mind: The evolutionary roots of human food cognition

rioux¹,

Wertz²,

Rumiati³

et al. 2022

Preprint

View full text Add to dashboard Cite

Research on food cognition has overlooked that our modern food environment strongly differs from the environment in which our ancestors lived. In fact, in most cultures today, many decisions about food are made during a trip to the grocery store, where we might ask ourselves: Are these canned tomatoes healthy? but certainly not: Are they edible or toxic? In contrast, our ancestors had to navigate their natural environment to Find and Evaluate the beneficial foods in that environment, while Excluding the costs of consuming something harmful, and finally Decide which entities to eat. We call this the FEED problem and present an innovative perspective on food cognition. We argue that a focus on FEED can bridge the gap between different findings in the field of food cognition and shed a new light on the cognitive and brain mechanisms that drive our food behaviors. It has the potential to promote novel intervention strategies, at a time when the rise of food-related health and environmental challenges calls for action.

show abstract

“…If this is the case, then cortical regions might be expected to respond differently to colored stimuli. (Rosenthal et al, 2018), for example, reported that the color-tuning of inferior temporal (IT) cortex correlates with the color statistics of parts of scenes labeled as objects, and color-biased regions appear to be selective for the colors of food (Pennock et al, 2022; Khosla et al, 2022). Differences in color preference across visual field eccentricity is also consistent with the hypothesis that color contributes differently across the visual field.…”

Section: Introductionmentioning

confidence: 99%

Color and Spatial Frequency Provide Functional Signatures of Retinotopic Visual Areas

Loggia

Duffield

Braunlich

et al. 2022

Preprint

View full text Add to dashboard Cite

The visual cortex contains multiple retinotopic representations of the visual field. A long-standing question of visual neuroscience concerns the role of these different regions in visual computations and perception. Differences in psychophysical properties have also been noted across the visual field, varying with eccentricity and with the upper and lower hemifields. Inspired by studies that used linear-systems-defined stimuli to characterize single neuron responses, we collected fMRI data from two macaque monkeys as they passively viewed gratings that varied in color, saturation, and spatial frequency. We apply a novel variant of multivariate linear discriminant analysis to discover combinations of these features that best separated visual cortex, along three dimensions of representation: eccentricity; classic retinotopic progression (V1, V2, V3, V3a, V4, MT+); and upper vs lower visual field. As expected from prior work, eccentricity across all visual areas was characterized by responses to color and spatial frequency: foveal regions, compared to peripheral regions, showed greater responses to colors not restricted to the S-cone-axis and high spatial frequency. Retinotopic visual areas were characterized by their responses to saturation: relative to anterior visual areas (V3-V4, V3a, MT+), posterior areas (V1-V2) were more responsive to increases in saturation. And upper-vs-lower representations were distinguished by responses to colors modulating along the Daylight locus: ventral areas showed relatively higher responses to colors associated with the Daylight locus (orange/blue) as compared to responses to achromatic luminance. Together, these results provide a data driven approach that recovers functional signatures of retinotopic organization using color and spatial frequency, providing clues toward the different roles that the components of retinotopic cortex play in visual perception.

show abstract

A highly selective response to food in human visual cortex revealed by hypothesis-free voxel decomposition

Cited by 48 publications

References 54 publications

A data-driven investigation of human action representations

A data-driven investigation of human action representations

FEED your mind: The evolutionary roots of human food cognition

Color and Spatial Frequency Provide Functional Signatures of Retinotopic Visual Areas

Contact Info

Product

Resources

About