A map of perceived speed at low- and high-contrast across the spatiotemporal frequency surface

Kevin R. Brooks, Kirsten L. Challinor

Research output: Contribution to journalMeeting abstract


Variations of perceived speed with spatial frequency (SF), temporal frequency (TF), and contrast have been known for many years. However, these effects have largely been studied in isolation, preventing comparison of the perceived speed of stimuli across the spatiotemporal frequency surface. We present the first systematic study to establish the perceived speed of high- and low-contrast (70% and 7%) stimuli across a broad range of Fourier parameters (SFs from 0.25-8c/deg; TFs from 2-16Hz; speeds from 0.25–64deg/s). In a temporal ‘yes-no’ speed discrimination paradigm, participants indicated whether a grating (standard: fixed speed) or a high-contrast dot pattern (test: variable speed) appeared faster. Points of subjective equality (PSEs) and just noticeable differences (JNDs) were established from the fitting of cumulative Gaussian functions. Test dot patterns had the same general form for all conditions of standard grating SF and TF, allowing for the comparison of PSEs and JNDs across Fourier space. Results showed that Weber fractions showed little variation over a broad range of spatiotemporal parameters and across contrast values. As in many previous reports, high-contrast stimuli appeared faster than low-contrast stimuli in the vast majority of cases, with some idiosyncratic exceptions. Generally, perceived speed increased with spatial frequency, particularly for low temporal frequency stimuli. The implications of these results for models of human speed perception are discussed.
Original languageEnglish
Pages (from-to)248-248
Number of pages1
Issue number4
Publication statusPublished - 2012
EventAVA Christmas Meeting - York, United Kingdom
Duration: 19 Dec 201119 Dec 2011


Dive into the research topics of 'A map of perceived speed at low- and high-contrast across the spatiotemporal frequency surface'. Together they form a unique fingerprint.

Cite this