Representing the location of manipulable objects in shape-selective occipitotemporal cortex:: beyond retinotopic reference frames?

Research output: Contribution to journalArticleResearchpeer-review

Abstract

When interacting with objects, we have to represent their location relative to our bodies. To facilitate bodily reactions, location may be encoded in the brain not just with respect to the retina (retinotopic reference frame), but also in relation to the head, trunk or arm (collectively spatiotopic reference frames). While spatiotopic reference frames for location encoding can be found in brain areas for action planning, such as parietal areas, there is debate about the existence of spatiotopic reference frames in higher-level occipitotemporal visual areas. In an extensive multi-voxel pattern analysis (MVPA) fMRI study using faces, headless bodies and scenes stimuli, Golomb and Kanwisher (2012) did not find evidence for spatiotopic reference frames in shape-selective occipitotemporal cortex. This finding is important for theories of how stimulus location is encoded in the brain. It is possible, however, that their failure to find spatiotopic reference frames is related to their stimuli: we typically do not manipulate faces, headless bodies or scenes. It is plausible that we only represent body-centred location when viewing objects that are typically manipulated. Here, we tested for object location encoding in shape-selective occipitotemporal cortex using manipulable object stimuli (balls and cups) in a MVPA fMRI study. We employed Bayesian analyses to determine sample size and evaluate the sensitivity of our data to test the hypothesis that location can be encoded in a spatiotopic reference frame in shape- selective occipitotemporal cortex over the null hypothesis of no spatiotopic location encoding. We found strong evidence for retinotopic location encoding consistent with previous findings that retinotopic reference frames are common neural representations of object location. In contrast, when testing for spatiotopic encoding, we found evidence that object location information for small manipulable objects is not decodable in relation to the body in shape-selective occipitotemporal cortex. Post-hoc exploratory analyses suggested that spatiotopic aspects might modulate retinotopic location encoding. Overall, our findings provide evidence that there is no spatiotopic encoding that is independent of retinotopic location in shape-selective occipitotemporal cortex.
LanguageEnglish
Pages132-150
Number of pages19
JournalCortex
Volume106
DOIs
Publication statusPublished - Sep 2018

Fingerprint

Brain
Magnetic Resonance Imaging
Bayes Theorem
Sample Size
Retina
Arm
Head

Cite this

@article{c6f26300bda940d5a3920c35275a75db,
title = "Representing the location of manipulable objects in shape-selective occipitotemporal cortex:: beyond retinotopic reference frames?",
abstract = "When interacting with objects, we have to represent their location relative to our bodies. To facilitate bodily reactions, location may be encoded in the brain not just with respect to the retina (retinotopic reference frame), but also in relation to the head, trunk or arm (collectively spatiotopic reference frames). While spatiotopic reference frames for location encoding can be found in brain areas for action planning, such as parietal areas, there is debate about the existence of spatiotopic reference frames in higher-level occipitotemporal visual areas. In an extensive multi-voxel pattern analysis (MVPA) fMRI study using faces, headless bodies and scenes stimuli, Golomb and Kanwisher (2012) did not find evidence for spatiotopic reference frames in shape-selective occipitotemporal cortex. This finding is important for theories of how stimulus location is encoded in the brain. It is possible, however, that their failure to find spatiotopic reference frames is related to their stimuli: we typically do not manipulate faces, headless bodies or scenes. It is plausible that we only represent body-centred location when viewing objects that are typically manipulated. Here, we tested for object location encoding in shape-selective occipitotemporal cortex using manipulable object stimuli (balls and cups) in a MVPA fMRI study. We employed Bayesian analyses to determine sample size and evaluate the sensitivity of our data to test the hypothesis that location can be encoded in a spatiotopic reference frame in shape- selective occipitotemporal cortex over the null hypothesis of no spatiotopic location encoding. We found strong evidence for retinotopic location encoding consistent with previous findings that retinotopic reference frames are common neural representations of object location. In contrast, when testing for spatiotopic encoding, we found evidence that object location information for small manipulable objects is not decodable in relation to the body in shape-selective occipitotemporal cortex. Post-hoc exploratory analyses suggested that spatiotopic aspects might modulate retinotopic location encoding. Overall, our findings provide evidence that there is no spatiotopic encoding that is independent of retinotopic location in shape-selective occipitotemporal cortex.",
keywords = "visual perception, action, object location encoding, manipulable objects, fMRI, occipitotemporal cortex",
author = "Regine Zopf and Marina Butko and Alexandra Woolgar and Mark Williams and Rich, {Anina N}",
year = "2018",
month = "9",
doi = "10.1016/j.cortex.2018.05.009",
language = "English",
volume = "106",
pages = "132--150",
journal = "Cortex",
issn = "0010-9452",
publisher = "Elsevier",

}

Representing the location of manipulable objects in shape-selective occipitotemporal cortex: beyond retinotopic reference frames? / Zopf, Regine; Butko, Marina; Woolgar, Alexandra; Williams, Mark; Rich, Anina N.

In: Cortex, Vol. 106, 09.2018, p. 132-150.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Representing the location of manipulable objects in shape-selective occipitotemporal cortex:

T2 - Cortex

AU - Zopf,Regine

AU - Butko,Marina

AU - Woolgar,Alexandra

AU - Williams,Mark

AU - Rich,Anina N

PY - 2018/9

Y1 - 2018/9

N2 - When interacting with objects, we have to represent their location relative to our bodies. To facilitate bodily reactions, location may be encoded in the brain not just with respect to the retina (retinotopic reference frame), but also in relation to the head, trunk or arm (collectively spatiotopic reference frames). While spatiotopic reference frames for location encoding can be found in brain areas for action planning, such as parietal areas, there is debate about the existence of spatiotopic reference frames in higher-level occipitotemporal visual areas. In an extensive multi-voxel pattern analysis (MVPA) fMRI study using faces, headless bodies and scenes stimuli, Golomb and Kanwisher (2012) did not find evidence for spatiotopic reference frames in shape-selective occipitotemporal cortex. This finding is important for theories of how stimulus location is encoded in the brain. It is possible, however, that their failure to find spatiotopic reference frames is related to their stimuli: we typically do not manipulate faces, headless bodies or scenes. It is plausible that we only represent body-centred location when viewing objects that are typically manipulated. Here, we tested for object location encoding in shape-selective occipitotemporal cortex using manipulable object stimuli (balls and cups) in a MVPA fMRI study. We employed Bayesian analyses to determine sample size and evaluate the sensitivity of our data to test the hypothesis that location can be encoded in a spatiotopic reference frame in shape- selective occipitotemporal cortex over the null hypothesis of no spatiotopic location encoding. We found strong evidence for retinotopic location encoding consistent with previous findings that retinotopic reference frames are common neural representations of object location. In contrast, when testing for spatiotopic encoding, we found evidence that object location information for small manipulable objects is not decodable in relation to the body in shape-selective occipitotemporal cortex. Post-hoc exploratory analyses suggested that spatiotopic aspects might modulate retinotopic location encoding. Overall, our findings provide evidence that there is no spatiotopic encoding that is independent of retinotopic location in shape-selective occipitotemporal cortex.

AB - When interacting with objects, we have to represent their location relative to our bodies. To facilitate bodily reactions, location may be encoded in the brain not just with respect to the retina (retinotopic reference frame), but also in relation to the head, trunk or arm (collectively spatiotopic reference frames). While spatiotopic reference frames for location encoding can be found in brain areas for action planning, such as parietal areas, there is debate about the existence of spatiotopic reference frames in higher-level occipitotemporal visual areas. In an extensive multi-voxel pattern analysis (MVPA) fMRI study using faces, headless bodies and scenes stimuli, Golomb and Kanwisher (2012) did not find evidence for spatiotopic reference frames in shape-selective occipitotemporal cortex. This finding is important for theories of how stimulus location is encoded in the brain. It is possible, however, that their failure to find spatiotopic reference frames is related to their stimuli: we typically do not manipulate faces, headless bodies or scenes. It is plausible that we only represent body-centred location when viewing objects that are typically manipulated. Here, we tested for object location encoding in shape-selective occipitotemporal cortex using manipulable object stimuli (balls and cups) in a MVPA fMRI study. We employed Bayesian analyses to determine sample size and evaluate the sensitivity of our data to test the hypothesis that location can be encoded in a spatiotopic reference frame in shape- selective occipitotemporal cortex over the null hypothesis of no spatiotopic location encoding. We found strong evidence for retinotopic location encoding consistent with previous findings that retinotopic reference frames are common neural representations of object location. In contrast, when testing for spatiotopic encoding, we found evidence that object location information for small manipulable objects is not decodable in relation to the body in shape-selective occipitotemporal cortex. Post-hoc exploratory analyses suggested that spatiotopic aspects might modulate retinotopic location encoding. Overall, our findings provide evidence that there is no spatiotopic encoding that is independent of retinotopic location in shape-selective occipitotemporal cortex.

KW - visual perception

KW - action

KW - object location encoding

KW - manipulable objects

KW - fMRI

KW - occipitotemporal cortex

UR - http://purl.org/au-research/grants/arc/DE140100499

UR - http://purl.org/au-research/grants/arc/FT170100105

UR - http://www.scopus.com/inward/record.url?scp=85048739897&partnerID=8YFLogxK

U2 - 10.1016/j.cortex.2018.05.009

DO - 10.1016/j.cortex.2018.05.009

M3 - Article

VL - 106

SP - 132

EP - 150

JO - Cortex

JF - Cortex

SN - 0010-9452

ER -