thompson (1982) perceived rate of mvt depends on contrast

7/28/2019 Thompson (1982) Perceived Rate of Mvt Depends on Contrast

1/4

Vi si on Res. Vo. 22, pp 317 10 380. 1982Printed in Great Britain0042-6989/82/030377-04SO3.0010Pergamon Press Ltd

PERCEIVED RATE OF MOVEMENT DEPENDSON CONTRAST

PETER THOMPSON*The Psychological Laboratory, University of Cambridge, Cambridge, England

(Received 28 September 1980; in reuisedform 12 August 1981)Abstract-The perceived rate of movement of sine wave gratings has been measured over a range ofcontrasts. At temporal frequencies below about 8 Hz a decrease in contrast reduces apparent velocity. At16 Hz a reduction in contrast increases perceived velocity.

INTRODUCTIONThe appearance of moving patterns at contrastthreshold has been examined by several investigators.Kulikowski and Tolhurst (1973) have reported thatfast-moving gratings of low spatial frequency are seento flicker at detection threshold whereas slowly mov-ing gratings of high spatial frequency appear to bestationary at detection threshold. They have sug-gested that two different sub-systems, movement andpattern channels, detect these two different stimuli-hence the difference in their appearance at threshold.Although it seems reasonable that at contrastthreshold only a single channel is responsible fordetecting a stimulus, once above threshold the activityof more than one channel may contribute to theappearance of a stimulus. The present experimentsexamine the appearance of moving gratings abovethreshold and demonstrate that the perceived velocityof such stimuli is greatly affected by their contrast.

EXPERIMENT 1: VELOCITY MATCHINGGRATINGS OF UNEQUAL CONTRAST

Methods and Result sMoving sine wave gratings were generated on the

screens of two oscilloscopes placed side by side. Eachscreen had dimensions of 6 x 4 deg, separated by1 deg. Subjects fixated an illuminated point betweenthe two screens and varied the rate of movement ofthe grating on the right hand screen (the match grat-ing) to match the rate of movement of the grating onthe left hand screen (the test grating). Fuller details ofthe apparatus and methods are given in Thompson(1981). The match grating was of 0.25 contrastthroughout, the velocity and contrast of the test grat-ing were set by the experimenter. Test grating con-trasts of 0.250.178, 0.125, 0.088, 0.063 and 0.045 wereused: each of these represents a 0.15 log unit stepfrom its neighbour.

* Present address: Department of Psychology, Univer-sity of York, York, England.

Eyemovement records obtained from a subjectviewing moving gratings in the apparatus show nosystematic variation with temporal frequency,(Thompson 1976). This is not unexpected. Murphy(1978) has shown that gratings moving behind astationary fixation point produce negligible eye driftwhich does not vary with the velocity of the grating.

To prevent the subject adapting to the stimuli, thegratings were presented for 2.5 set, in which time hecould be expected to make only a very approximatevelocity match. Each presentation was followed by17.5 set during which both screens were maintained attheir mean luminance (31.6 cd m-) without any grat-ing present. During successive presentations of thegratings the subject was able to improve the velocitymatch until he was satisfied that the gratingsappeared to be moving at equal velocity. The matchgrating velocity was then recorded and a new teststimulus presented.

When the test grating contrast was 0.25 the sub-jects task was simply to match gratings of equal con-trast for velocity-these matches were regarded asbaseline measures. In all other conditions the contrastof the test grating was less than that of the matchgrating. The experimental results plot the perceivedvelocity of the test gratings as a fraction of their per-ceived velocity when at equal contrast with the matchgrating. A velocity match of less than 1.0 indicatesthat the effect of reducing the test grating contrast isto reduce the perceived velocity while a value greaterthan 1.0 indicates that a reduction in test grating con-trast increases its perceived velocity.

In Experiment 1 all gratings were of 2 c/deg spatialfrequency. Five different test grating velocities wereinvestigated, which at the six contrast levels givenabove produced 30 test stimuli. Each stimulus waspresented at least four times and all were in randomorder. The results of this experiment are presented inFig. 1. The results are clear. At temporal frequenciesbelow about 8 Hz perceived velocity is reduced ascontrast is reduced. At the highest temporal frequencyinvestigated, 16 Hz, there is a marked overestimationof speed as contrast is reduced. Campbell and Maffei

377


2/4

378 PETER THOMPSON

>.-

PGT 2 c/degP

> 1.4- RN c/deg1.3 -1.2 -I.1 -

I.0 -0.9 -0.6 -

I I I I II 2 4 6 16

Test grating rate (HZ)

EXPERIMENT 2: VELOCITY ESTIMATIONOF GRATINGS OF UNEQUAL

CONTRASTMethod and Results

In Experiment 2 subjects fixated between the twoscreens as before but stimuli were presented only onthe left hand screen. Two naive subjects werepresented with 2c/deg gratings of various contrastsmoving at various speeds and were asked to estimatetheir perceived rate of movement. Prior to the experi-ment and periodically during it. the subjects werepresented with a 10 Hz, 2 c/deg grating of 0.25 con-trast which they were told to regard as a standardvelocity of speed 10, with a stationary grating rep-resenting speed 0. Gratings of five rates of movement1. 2, 4, 8 and 16 Hz, and three contrast levels 0.25,0.15 and 0.1 contrast were presented in random order.The speed estimations of the 0.25 contrast gratingswere regarded as a baseline measure, with the lowercontrast gratings speed estimations being expressedas a fraction of these baselines. The results are shownin Fig. 3; their similarity to the results shown in Fig. 1indicates that the dependence of perceived velocity

Fig. 1. Velocity matching at different contrasts. All grat-ings at a constant spatial frequency of 2 c/deg. Subject con-trolled grating of 0.25 contrast. Test grating contrasts of:0.178 (0); 0.125 (x); 0.088 (A): 0.063 (0): 0.045 (0).Subject P.G.T. Subject R.N. RN I c/dogI.2

(1981) have recently reported that, at very slow ratesof rotary motion, increasing contrast increases per-ceived velocity, a result which concurs with thepresent finding.

For both subjects the changeover from underesti- RN 4 c/drgmation to overestimation of velocity occurs around a iltemporal frequency of 8 Hz, which for 2 c/deg gratingsis a velocity of 4 deg/sec. By repeating the experimentat other spatial frequencies we can discover whetherthis changeover depends on the temporal frequencyor the velocity of the gratings. Figure 2 shows theresults for gratings of 1, 4 and 8c/deg. In each casethe crossover from underestimation to overestimationoccurs at a constant temporal frequency, around 1.4 - RN 8 c/dog8 Hz. and not at a constant velocity. I.3 -

It could be argued that some aspects of these 1.2 -results reflect an artefact in the matching method I.1 -employed. Suppose that, as well as the perceived vel- 1.0 -ocity of the stimulus depending on its contrast, the 0.9 -perceived contrast of a stimulus depends on its vel- 0.6 -ocity. Then, if several different velocities of the match 0.7 -grating are equally acceptable as a velocity match to I I I I Ithe test grating, the subject might choose that velocity I 2 4 6 16which most nearly matched the test grating for con- Test grating rate (Hz)trast. A second experiment was carried out using anestimation technique to demonstrate that the results Fig. 2. Velocity matching at different contrasts. All grat-of Experiment 1 are not contaminated by any effect of ings at a constant spatial frequency of: (a) 1c/deg.(b) 4 c/deg. (c) 8 c/deg. Other details as in Fig. I.velocity on perceived contrast. Subject R.N.


3/4

Perceived rate of movement depends on contrast 379

1.2 -

0.0 -zuc 0.6-: 0.4 -

0.6 tI 1 I II 2 4 0 I6

Test grating rat0 (Hz)Fig. 3. Velocity magnitude estimations at different con-trasts. All gratings at a constant spatial frequency of2c/deg. Grating velocities estimated at 3 contrast levels.Contrasts of 0.16 (0) and 0.1 (A) expressed as a fractionof their perceived velocity at 0.25 contrast. Subject J.M.Subject P.P.

upon contrast cannot be explained as an artefact ofthe matching procedure used in Experiment 1.

DISCUSSION

Perceived velocity clearly depends on contrast, butthe mechanisms underlying this dependence remainobscure. The crossover from underestimation to over-estimation around 8 Hz suggests that the temporalfrequency rather than the velocity is the importantstimulus parameter. One test of this might be toexamine the perceived flicker rate of a uniform field asa function of temporal contrast. If similar results tothose reported here were found then the presentresults could be seen as the dependence solely of tem-poral frequency upon contrast which gives rise to dis-tortions in perceived velocity. Unfortunately it isalmost impossible to make any kind of sensible orconsistent match of temporal frequency between flick-ering fields of unequal contrast. I have tried and havefailed and both Harris and Sidwell (personal com-munications) have also attempted to carry out similarmeasurements, without great success. Perhaps an ex-periment which attempted to measure flicker rate insome other way, (e.g. the cross-modal matching withauditory frequency flutter employed by Fukuda 1977)could provide an answer.

It seems most likely, then, that the effect reportedhere is neither a purely spatial nor a purely temporalphenomenon. It could, however, be understood interms of a simple two channel model comprising aslow channel, which is most sensitive to stationaryand slow-moving patterns, and a fast channel sensi-tive to fast-moving patterns, (Fig. 4). At suprathresh-old levels the perceived velocity of stimuli would bedetermined by the relative activity in these channels.;this is by no means a new idea, having been discussedin general terms by Tolhurst et ul. (1973). Supposethat below about 8 Hz the slow channel has a lowercontrast threshold than the fast channel. Then reduc-ing the contrast of a slowly moving stimulus will biasthe relative activity in these channels, the slow chan-nel becoming relatively more sensitive to the stimulusuntil at very low contrasts the stimulus will bedetected by the slow channel alone. Therefore if per-ceived velocity depends upon the relative activity inthese two channels, reducing the contrast of a slowlymoving pattern should reduce its apparent speed,which is the result found. The same argument canexplain the increase in velocity of fast moving stimuliwhen their contrast is reduced.

The results reported in this paper could also be the The experimental data suggest that there cannot beconsequence of a dependence of spatial frequency an exact correspondence between the two channelsupon contrast. Georgeson (1980) has reported large described here and the pattern and movement chan-increases in the perceived spatial frequency of station- nels described by Kulikowski and Tolhurst (1973) andary gratings when their contrast was decreased. In others. First. the switch from velocity underestimationExperiment 1, therefore, subjects were matching the to overestimation has been found to be independent

perceived velocity of gratings which, while havingequal real spatial frequencies, had different apparentspatial frequencies. Could this explain the reductionsin perceived velocity found in Experiment l? George-son argues as follows: suppose that spatial frequencychannels are labelled, that is their spatial significanceis invariant; now. if the optimal spatial frequency fora channel deceases as contrast is reduced, then at lowcontrast the optimal stimulus for a channel will be aspatial frequency lower than its labelled value. Forexample, a channel labelled 4 c/deg will be optimallystimulated by a grating of, say, 3c/deg at low con-trast. In other words, that 3 c/deg grating optimallyexcites a channel labelled as 4 c/deg and will thereforebe perceived as a 4 c/deg grating. We can apply thisline of argument to the present results. If the per-ceived velocity of a grating were determined by somecomparison of its component temporal and spatialfrequencies, then a dependence of perceived velocityupon contrast could be regarded as a purely spatialphenomenon. As contrast is reduced, perceived spatialfrequency increases and, if temporal frequencyremains unaffected, perceived velocity must decrease.

If such an explanation were to explain all thepresent data it would be necessary that as temporalfrequency increased up to about 8 Hz the effects ofcontrast reduction on perceived spatial frequencywould decrease and that at higher temporal frequen-cies contrast reduction would actually decrease per-ceived spatial frequency. There is no evidence for this.


4/4

380 PETER THOMPSONconfused changes in perceived velocity brought aboutby the adaptation to movement per use with changesin perceived velocity brought about by the contrastreducing effect of the adaptation stimulus, (Blakemorecf ul., 1973). Interestingly, whereas the changes in per-ceived velocity resulting from contrast reductionappear temporal frequency tuned, the velocity after-effects resulting from adaptation to movement are vel-ocity tuned (Thompson 1981).

Temporal frequencyFig. 4. A mechanism of the dependence of perceived vel-ocity upon apparent contrast. Perceived velocity is deter-mined by the relative activity in the two channels. It isassumed that both channels respond in the same way tochanges in contrast. At low temporal frequencies. wherethe slow channel is more sensitive, reducing stimulus con-trast will reduce the relative contribution of the fast chan-nel, (and therefore the perceived rate of movement), until atvery low contrast levels the stimulus is detected by the slowchannel alone. At high temporal frequencies the position isreversed, reductions in contrast reducing the relative con-tribution of the slow channels response. The crossoverfrom the slow channel being more sensitive to the fastchannel being more sensitive appears to be around 8 Hz.under the conditions presented here.

of spatial frequency, whereas the relative sensitivity ofpattern and movement channels changes considerablyover spatial frequency. Secondly, the change-overfrom underestimation to overestimation should havebeen around 24Hz and not around 8-10Hz asfound.

That perceived velocity is dependent upon thestimulus contrast has important implications for ourvelocity judgements in the real world (e.g. driving infog) as well as in laboratory experiments. As anexample of the latter it might be noted that experi-ments on velocity aftereffects (the change in velocityobserved following adaptation to movement) have

Acknowledgements-This work was carried out when theauthor was supported by a studentship from the S.S.R.C.Some apparatus was loaned by Professor Cohn Blakemore(M.R.C. Grant G972/463/B). Preparation of this manu-script was partly supported by an S.R.C. grant(GRIA73817) to the author. I thank 01 Braddick, DaveTolhurst and Dave Heeley for their encouragement. andMike Harris and Andrew Sidwell for valuable discussionsof their own work in this area prior to publication

REFERENCESBlakemore C., Muncey J. P. J. and Ridley R. M. (1973)Stimulus specificity in the human visual system. Vision

Res. 13, 1915-1931.Campbell F. W. and Maffei L. (1981) The influence of spa-tial frequency and contrast on the perception of movingpatterns. Vision Res. 21, 713-721.Fukuda T. (1977) Subjective frequency in flicker percep-tion. Percept. Motor Skis 45, 203-210.Georgeson M. A. (1980) Spatial frequency analysis in earlyvisual processing. Phil. Trans. R. Sot. Ser. B 290, 1 -22,Kulikowski J. and Tolhurst D. J. (1973) Psychophysicalevidence for sustained and transient neurones in thehuman visual system. 1. Physiol. 232, 149-162.Murphy B. J. (1978) Pattern thresholds for moving andstationary gratings. Vision Res. 18, 521-530.Thompson P. G. (1976) Velocity after-effects and the per-ception of movement. Ph.D. thesis, University of Cam-bridge, England.Thompson P. (1980) Velocity aftereffects: the effects ofadaptation to moving stimuli on the perception of subse-quently seen moving stimuli. vision Res. 21, 337-345.Tolhurst D. J., Sharpe C. R. and Hart G. (1973) The analy-sis of the drift rate of moving sinusoidal gratings. Vision

Res. 13, 2545-2555.

thompson (1982) perceived rate of mvt depends on contrast

Documents