Vision Research 39 (1999) 3721–3727
Perception of coherent motion, biological motion and
form-from-motion under dim-light conditions
E.D. Grossman, R. Blake *
Department of Psychology, Vanderbilt Vision Research Center, Vanderbilt Uni?ersity, Nash?ille, TN 37240, USA
Received 19 August 1998; received in revised form 13 March 1999
Three experiments investigated several aspects of motion perception at high and low luminance levels. Detection of weak
coherent motion in random dot cinematograms was unaffected by light level over a range of dot speeds. The ability to judge form
from motion was, however, impaired at low light levels, as was the ability to discriminate normal from phase-scrambled biological
motion sequences. The difficulty distinguishing differential motions may be explained by increased spatial pooling at low light
levels. © 1999 Elsevier Science Ltd. All rights reserved.
Keywords: Motion perception; Light level; Form from motion; Biological motion; Coherence; Random dot cinematogram
As everyone knows, it’s hard to see when it’s dark.
Colors fade to shades of gray, stereoscopic depth per-
ception deteriorates, and reading can become impossi-
ble because of reduced visual acuity. These changes in
visual performance under dim-light conditions are well
explained by changes in the visual mode of processing;
a shift from cone- to rod-dominated photoreception,
and changes in the balance between center/surround
mechanisms of retinal ganglion cells effectively enlarge
the cells’ summation area (Barlow, Fitzhugh & Kuffler,
1957; Derrington & Lennie, 1982). In general, at lower
light levels spatial resolution is compromised in the
interests of sensitivity.
However, based simply on experience one is not
aware of wholesale changes in the ability to see object
movement under dim-light conditions, nor does our
reliance on optic flow for navigation seem seriously
hampered. Yet mechanisms responsible for motion per-
ception receive inputs from the same ‘front-end’ mecha-
nisms whose response properties adversely affect vision
at low light levels. Moreover, it is well established that
the temporal response of the visual system becomes
more sluggish at low light levels (e.g. Matin, 1968),
which can be modeled as a blurring of the temporal
impulse response (Kelly, 1971). To the extent that early
temporal filters are involved in the analysis of motion
information, one would reasonably expect reductions in
light level to impact perception of motion. Much recent
research has been devoted to describing the human
capacity to perceive motion and to understanding the
neural mechanisms involved, but the vast majority of
that work has been limited to perception at high lumi-
nances. Only a handful of studies have assessed motion
perception at low light levels (e.g. Dawson & Di Lollo,
1990) and none of those has examined more refined
aspects of motion perception such as form from
Accordingly, this paper compares three aspects of
motion perception–coherence detection, form from
motion (FFM) and biological motion-at high and low
light levels. From a computational standpoint, these
motion tasks would seem to involve different processing
operations. Detection of coherent motion requires inte-
gration of motion signals over space and time, while
FFM and biological motion require spatial and tempo-
ral differentiation of motion signals. Adding to its
complexity, biological motion entails dynamic, hierar-
chically arranged pendular motions which, when viewed
* Corresponding author. Tel.: +1-615-343-7010; fax: +1-615-343-
E-mail address: firstname.lastname@example.org (R. Blake)
0042-6989/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
E.D. Grossman, R. Blake / Vision Research 39 (1999) 3721–37273727
differential activation among neurons registering mo-
tion in nearby regions of visual space. For example, we
would expect impairment within processes involved in
segregation of one cluster of motion vectors from a
background of different vectors, for pooling could blur
those motion boundaries. This kind of process, of
course, would be required for perception of our FFM
stimulus. Motion boundaries may also be highlighted in
virtue of the motion opponency described for receptive
fields of neurons in the middle temporal visual area in
monkey, where opposite directions of motion in adja-
cent regions of the visual field generate particularly
strong responses (Allman, Miezin & McGuinness,
1985). If the balance between these opponent processes
shifts with dark adaptation—in a manner comparable
to the shifts seen in retinal ganglion cells (Barlow et al.,
1957)—motion boundaries would be blurred. To the
extent that perception of biological motion depends on
spatial relations among relevant motion tokens, spatial
pooling also could be responsible for generally poorer
performance on this task at low luminance levels.
Finally, it is worth considering our results in light of
work by Purpura, Kaplan and Shapley (1988) who
measured contrast gain control in parvo- and magno-
cellular retinal ganglion cells at different levels of light
adaptation. At mesopic and scotopic luminance levels,
the responses of P cells were severely reduced, rendering
them almost ‘blind’ under these conditions; M neurons,
in contrast, maintained high levels of responsiveness.
Now to the extent that M cells provide the gateway to
motion mechanisms in the brain as commonly believed,
we would expect motion perception to survive large
reductions in light level. Our results show that this is
the case, although observers do experience more
difficulty extracting shape from those motion signals
and more difficulty assembling local motion signals into
globally coherent biological events. In a sense, motion
is easy to see at low light levels, but global spatial
structure carried by the motion is not. One might
therefore construe the impairments in performance on
biological motion and FFM as implicating P-pathway
involvement in those tasks. It is important to keep in
mind, however, that performance on those tasks was
possible at the low light level, albeit with decreased
efficiency. Any conclusions involving relative activation
of M and P channels on FFM and biological motion
tasks will require testing under other conditions
thought to isolate these two pathways.
This work was supported by grants from NIH
(EY07760 and P30-EY08126). EG is a NIH Predoc-
toral Trainee. We thank Sang-Hun Lee for helpful
discussion and Eric Hiris for help with programming.
Ahlstro ¨m, V., Blake, R., & Ahlstro ¨m, U. (1997). Perception of
biological motion. Perception, 26, 1539–1548.
Allman, J. M., Miezin, R., & McGuinness, E. (1985). Direction- and
velocity-specific responses from beyond the classical receptive field
in the middle temporal visual area (MT). Perception, 14, 105–126.
Barlow, H. B., Fitzhugh, R., & Kuffler, S. W. (1957). Changes in
organization of the receptive fields of the cat’s retina during dark
adaptation. Journal of Physiology, 137, 327–337.
Bertenthal, B. I., & Pinto, J. (1994). Global processing of biological
motions. Psychological Science, 5, 221–225.
Britten, K. H., Newsome, S. T., Shadlen, M. N., Celebrini, S., &
Movshon, J. A. (1996). A relationship between behavioral choice
and the visual responses of neurons in macaque MT. Visual
Neuroscience, 13, 87–100.
Dawson, M., & Di Lollo, V. (1990). Effects of adapting luminance and
stimulus contrast on the temporal and spatial limits of short-range
motion. Vision Research, 30, 415–429.
Derrington, A. M., & Lennie, P. (1982). The influence of temporal
frequency and adaptation level on receptive field organization of
Johansson, G. (1973). Visual perception of biological motion and a
model for its analysis. Perception and Psychophysics, 14, 201–211.
Kelly, D. H. (1971). Theory of flicker and transient responses, II.
Counterphase gratings. Journal of the Optical Society of America,
Maloney, L. T. (1990). Confidence intervals for the parameters of
psychometric functions. Perception and Psychophysics, 47, 127–
Matin, L. (1968). Critical duration, the differential luminance
threshold, critical flicker frequency, and visual adaptation: a
theoretical treatment. Journal of the Optical Society of America, 58,
Mayser, H., Eckle, T., Braun, D. I., Gegenfurtner, K. R., & Sharpe,
L. T. (1998). Motion perception at scotopic levels (abstract).
In?estigati?e Ophthalmology and Visual Science, 39, 4973.
Milner, A. D., & Jeeves, M. A. (1985). Visual analysis of body
movements by neurones in the temporal cortex of the macaque
monkey: a preliminary report. Beha?ioural Brain Research, 16,
Oram, M. W., & Perrett, D. I. (1994). Responses of anterior superior
temporal polysensory (STPa) neurons to ‘biological motion’ stim-
uli. Journal of Cogniti?e Neuroscience, 6, 99–116.
Purpura, K., Kaplan, E., & Shapley, R. M. (1988). Background light
and the contrast gain of primate P and M retinal ganglion cells.
Proceedings of the National Academy of Science, 85, 4534–4537.
Schenk, T., & Zihl, J. (1997a). Visual motion perception after brain
damage: I. Deficits in global motion perception. Neuropsychologia,
Schenk, T., & Zihl, J. (1997b). Visual motion perception after brain
damage: II. Deficits in form-from-motion perception. Neuropsy-
chologia, 35(9), 1299–1310.
A computational analysis of the relationship between neuronal and
behavioral responses to visual motion. Journal of Neuroscience, 16,
Sumi, S. (1984). Upside down presentation of the Johansson moving
light spot pattern. Perception, 13, 283–286.
Takeuchi, T., & De Valois, K. K. (1997). Motion-reversal reveals two
motion mechanisms functioning in scotopic vision. Vision Research,
Vaina, L. M., Lemay, M., Bienfang, D. C., Choi, A. Y., & Nakayama,
K. (1990). Intact ‘biological motion’ and ‘structure from motion’
perception in a patient with impaired motion mechanisms: a case
study. Visual Neuroscience, 5, 353–370.
from stochastic local motions. Vision Research, 24, 55–62.