Model fitting versus curve fitting: A model of renormalization provides a better account of age aftereffects than a model of local repulsion

S.F. O'Neil, A. Mac, Gillian Rhodes, M.A. Webster

Research output: Contribution to journalArticle

83 Downloads (Pure)

Abstract

© The Author(s) 2015. Recently, we proposed that the aftereffects of adapting to facial age are consistent with a renormalization of the perceived age (e.g., so that after adapting to a younger or older age, all ages appear slightly older or younger, respectively). This conclusion has been challenged by arguing that the aftereffects can also be accounted for by an alternative model based on repulsion (in which facial ages above or below the adapting age are biased away from the adaptor). However, we show here that this challenge was based on allowing the fitted functions to take on values which are implausible and incompatible across the different adapting conditions. When the fits are constrained or interpreted in terms of standard assumptions about normalization and repulsion, then the two analyses both agree in pointing to a pattern of renormalization in age aftereffects.
Original languageEnglish
Article number 2041669515613669
Pages (from-to)1-7
Number of pages7
Journali-Perception
Volume6
Issue number6
DOIs
Publication statusPublished - 18 Nov 2015

Fingerprint

Curve fitting

Cite this

@article{89cbf16ededb4bf8a4747a20cd9dd544,
title = "Model fitting versus curve fitting: A model of renormalization provides a better account of age aftereffects than a model of local repulsion",
abstract = "{\circledC} The Author(s) 2015. Recently, we proposed that the aftereffects of adapting to facial age are consistent with a renormalization of the perceived age (e.g., so that after adapting to a younger or older age, all ages appear slightly older or younger, respectively). This conclusion has been challenged by arguing that the aftereffects can also be accounted for by an alternative model based on repulsion (in which facial ages above or below the adapting age are biased away from the adaptor). However, we show here that this challenge was based on allowing the fitted functions to take on values which are implausible and incompatible across the different adapting conditions. When the fits are constrained or interpreted in terms of standard assumptions about normalization and repulsion, then the two analyses both agree in pointing to a pattern of renormalization in age aftereffects.",
author = "S.F. O'Neil and A. Mac and Gillian Rhodes and M.A. Webster",
year = "2015",
month = "11",
day = "18",
doi = "10.1177/2041669515613669",
language = "English",
volume = "6",
pages = "1--7",
journal = "i-Perception",
issn = "2041-6695",
publisher = "Pion Ltd",
number = "6",

}

Model fitting versus curve fitting: A model of renormalization provides a better account of age aftereffects than a model of local repulsion. / O'Neil, S.F.; Mac, A.; Rhodes, Gillian; Webster, M.A.

In: i-Perception, Vol. 6, No. 6, 2041669515613669, 18.11.2015, p. 1-7.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Model fitting versus curve fitting: A model of renormalization provides a better account of age aftereffects than a model of local repulsion

AU - O'Neil, S.F.

AU - Mac, A.

AU - Rhodes, Gillian

AU - Webster, M.A.

PY - 2015/11/18

Y1 - 2015/11/18

N2 - © The Author(s) 2015. Recently, we proposed that the aftereffects of adapting to facial age are consistent with a renormalization of the perceived age (e.g., so that after adapting to a younger or older age, all ages appear slightly older or younger, respectively). This conclusion has been challenged by arguing that the aftereffects can also be accounted for by an alternative model based on repulsion (in which facial ages above or below the adapting age are biased away from the adaptor). However, we show here that this challenge was based on allowing the fitted functions to take on values which are implausible and incompatible across the different adapting conditions. When the fits are constrained or interpreted in terms of standard assumptions about normalization and repulsion, then the two analyses both agree in pointing to a pattern of renormalization in age aftereffects.

AB - © The Author(s) 2015. Recently, we proposed that the aftereffects of adapting to facial age are consistent with a renormalization of the perceived age (e.g., so that after adapting to a younger or older age, all ages appear slightly older or younger, respectively). This conclusion has been challenged by arguing that the aftereffects can also be accounted for by an alternative model based on repulsion (in which facial ages above or below the adapting age are biased away from the adaptor). However, we show here that this challenge was based on allowing the fitted functions to take on values which are implausible and incompatible across the different adapting conditions. When the fits are constrained or interpreted in terms of standard assumptions about normalization and repulsion, then the two analyses both agree in pointing to a pattern of renormalization in age aftereffects.

U2 - 10.1177/2041669515613669

DO - 10.1177/2041669515613669

M3 - Article

VL - 6

SP - 1

EP - 7

JO - i-Perception

JF - i-Perception

SN - 2041-6695

IS - 6

M1 - 2041669515613669

ER -