Spiral mechanisms are required to account for summation of complex motion components

T. S. Meese*, S. J. Anderson

*Corresponding author for this work

Research output: Contribution to journalMeeting abstract

Abstract

Stimuli from one family of complex motions are defined by their spiral pitch, where cardinal axes represent signed expansion (pitch = 0° & 180°) and signed rotation (pitch = -90° & 90°). Intermediate spirals are represented by intermediate pitches. It is well established that vision contains mechanisms that sum over space and direction to detect these stimuli (Morrone et al, 1995, Nature, 376, 507-509; Harris & Meese, 1996, Perception, 25-supp, 129 and one intriguing possibility is that 4 cardinal mechanisms encode the entire family. We extended earlier work using a 2IFC technique and subthreshold summation to investigate this possibility. Subjects had to discriminate between random dot kinematograms constructed from 'signal+noise' in one interval and 'noise' alone in the other interval, where 'signal' comprised two components (A, B). Stimuli had a diameter of 5°, a duration of 300ms and contained 440 randomly positioned dots that traveled linearly through 10% of their distances from the origin. In the 'noise' interval, all dots had their directions randomized individually and in the 'signal+noise' interval, a staircase controlled the percentage of 'signal' dots. In Exp. 1, component A had a pitch of 0°, and in Exp. 2 it had a pitch of -45°. In both experiments, the pitch of component B was varied in 15° steps from 0° to 90° relative to the pitch of component A. Results were similar in both experiments with summation between the components decreasing as a function of pitch difference. Models contained mechanisms spaced at a pitch of either 90° (cardinal model) or 45° (spiral model). Probability summation was assumed between mechanisms and direction half-bandwidth (bw) was fit by the simplex algorithm. Although the cardinal model could account for the results of Exp. 1 (bw=69°), only the spiral model could account for the results of both exps. (bw=47°). We conclude that the detection of complex motion in human vision requires both cardinal and spiral mechanisms.

Original languageEnglish
JournalJournal of Vision
Volume1
Issue number3
DOIs
Publication statusPublished - 2001

Fingerprint

Noise
Direction compound

Cite this

@article{2b429889e53543388f57ca4b8ddb6ad8,
title = "Spiral mechanisms are required to account for summation of complex motion components",
abstract = "Stimuli from one family of complex motions are defined by their spiral pitch, where cardinal axes represent signed expansion (pitch = 0° & 180°) and signed rotation (pitch = -90° & 90°). Intermediate spirals are represented by intermediate pitches. It is well established that vision contains mechanisms that sum over space and direction to detect these stimuli (Morrone et al, 1995, Nature, 376, 507-509; Harris & Meese, 1996, Perception, 25-supp, 129 and one intriguing possibility is that 4 cardinal mechanisms encode the entire family. We extended earlier work using a 2IFC technique and subthreshold summation to investigate this possibility. Subjects had to discriminate between random dot kinematograms constructed from 'signal+noise' in one interval and 'noise' alone in the other interval, where 'signal' comprised two components (A, B). Stimuli had a diameter of 5°, a duration of 300ms and contained 440 randomly positioned dots that traveled linearly through 10{\%} of their distances from the origin. In the 'noise' interval, all dots had their directions randomized individually and in the 'signal+noise' interval, a staircase controlled the percentage of 'signal' dots. In Exp. 1, component A had a pitch of 0°, and in Exp. 2 it had a pitch of -45°. In both experiments, the pitch of component B was varied in 15° steps from 0° to 90° relative to the pitch of component A. Results were similar in both experiments with summation between the components decreasing as a function of pitch difference. Models contained mechanisms spaced at a pitch of either 90° (cardinal model) or 45° (spiral model). Probability summation was assumed between mechanisms and direction half-bandwidth (bw) was fit by the simplex algorithm. Although the cardinal model could account for the results of Exp. 1 (bw=69°), only the spiral model could account for the results of both exps. (bw=47°). We conclude that the detection of complex motion in human vision requires both cardinal and spiral mechanisms.",
author = "Meese, {T. S.} and Anderson, {S. J.}",
year = "2001",
doi = "10.1167/1.3.160",
language = "English",
volume = "1",
journal = "Journal of Vision",
issn = "1534-7362",
publisher = "Association for Research in Vision and Ophthalmology Inc.",
number = "3",

}

Spiral mechanisms are required to account for summation of complex motion components. / Meese, T. S.; Anderson, S. J.

In: Journal of Vision, Vol. 1, No. 3, 2001.

Research output: Contribution to journalMeeting abstract

TY - JOUR

T1 - Spiral mechanisms are required to account for summation of complex motion components

AU - Meese, T. S.

AU - Anderson, S. J.

PY - 2001

Y1 - 2001

N2 - Stimuli from one family of complex motions are defined by their spiral pitch, where cardinal axes represent signed expansion (pitch = 0° & 180°) and signed rotation (pitch = -90° & 90°). Intermediate spirals are represented by intermediate pitches. It is well established that vision contains mechanisms that sum over space and direction to detect these stimuli (Morrone et al, 1995, Nature, 376, 507-509; Harris & Meese, 1996, Perception, 25-supp, 129 and one intriguing possibility is that 4 cardinal mechanisms encode the entire family. We extended earlier work using a 2IFC technique and subthreshold summation to investigate this possibility. Subjects had to discriminate between random dot kinematograms constructed from 'signal+noise' in one interval and 'noise' alone in the other interval, where 'signal' comprised two components (A, B). Stimuli had a diameter of 5°, a duration of 300ms and contained 440 randomly positioned dots that traveled linearly through 10% of their distances from the origin. In the 'noise' interval, all dots had their directions randomized individually and in the 'signal+noise' interval, a staircase controlled the percentage of 'signal' dots. In Exp. 1, component A had a pitch of 0°, and in Exp. 2 it had a pitch of -45°. In both experiments, the pitch of component B was varied in 15° steps from 0° to 90° relative to the pitch of component A. Results were similar in both experiments with summation between the components decreasing as a function of pitch difference. Models contained mechanisms spaced at a pitch of either 90° (cardinal model) or 45° (spiral model). Probability summation was assumed between mechanisms and direction half-bandwidth (bw) was fit by the simplex algorithm. Although the cardinal model could account for the results of Exp. 1 (bw=69°), only the spiral model could account for the results of both exps. (bw=47°). We conclude that the detection of complex motion in human vision requires both cardinal and spiral mechanisms.

AB - Stimuli from one family of complex motions are defined by their spiral pitch, where cardinal axes represent signed expansion (pitch = 0° & 180°) and signed rotation (pitch = -90° & 90°). Intermediate spirals are represented by intermediate pitches. It is well established that vision contains mechanisms that sum over space and direction to detect these stimuli (Morrone et al, 1995, Nature, 376, 507-509; Harris & Meese, 1996, Perception, 25-supp, 129 and one intriguing possibility is that 4 cardinal mechanisms encode the entire family. We extended earlier work using a 2IFC technique and subthreshold summation to investigate this possibility. Subjects had to discriminate between random dot kinematograms constructed from 'signal+noise' in one interval and 'noise' alone in the other interval, where 'signal' comprised two components (A, B). Stimuli had a diameter of 5°, a duration of 300ms and contained 440 randomly positioned dots that traveled linearly through 10% of their distances from the origin. In the 'noise' interval, all dots had their directions randomized individually and in the 'signal+noise' interval, a staircase controlled the percentage of 'signal' dots. In Exp. 1, component A had a pitch of 0°, and in Exp. 2 it had a pitch of -45°. In both experiments, the pitch of component B was varied in 15° steps from 0° to 90° relative to the pitch of component A. Results were similar in both experiments with summation between the components decreasing as a function of pitch difference. Models contained mechanisms spaced at a pitch of either 90° (cardinal model) or 45° (spiral model). Probability summation was assumed between mechanisms and direction half-bandwidth (bw) was fit by the simplex algorithm. Although the cardinal model could account for the results of Exp. 1 (bw=69°), only the spiral model could account for the results of both exps. (bw=47°). We conclude that the detection of complex motion in human vision requires both cardinal and spiral mechanisms.

UR - http://www.scopus.com/inward/record.url?scp=4143136597&partnerID=8YFLogxK

UR - https://jov.arvojournals.org/article.aspx?articleid=2119750

U2 - 10.1167/1.3.160

DO - 10.1167/1.3.160

M3 - Meeting abstract

AN - SCOPUS:4143136597

VL - 1

JO - Journal of Vision

JF - Journal of Vision

SN - 1534-7362

IS - 3

ER -