Ocular tracking as a measure of auditory motion perception

Abstract

Motion is a potent sub-modality of vision. Motion cues alone can be used to segment images into figure and ground and break camouflage. Specific patterns of motion support vivid percepts of form, guide locomotion by specifying directional heading and the passage of objects, and in case of an impending collision, the time to impact. Visual motion also drives smooth pursuit eye movements (SPEMs) that serve to stabilize the retinal image of objects in motion.

In contrast, the auditory system does not appear to be particularly sensitive to motion. We review the ambiguous status of auditory motion processing from the psychophysical and electrophysiological perspectives. We then report the results of two experiments that use ocular tracking performance as an objective measure of the perception of auditory motion in humans. We examine ocular tracking of auditory motion, visual motion, combined auditory + visual motion and imagined motion in both the frontal plane and in depth. The results demonstrate that ocular tracking of auditory motion is no better than ocular tracking of imagined motion. These results are consistent with the suggestion that, unlike the visual system, the human auditory system is not endowed with low-level motion sensitive elements. We hypothesize however, that auditory information may gain access to a recently described high-level motion processing system that is heavily dependent on `top-down' influences, including attention.

Introduction

The goal of perceptual processing is to create an internal (neural) representation of the external environment. A primary dimension of most sensory representations is concerned with the specification of spatial location. This is certainly the case for vision and audition, for it is indeed difficult to imagine a visual or auditory percept that does not include a localization component.

A great deal of neural circuitry is devoted to performing the computations needed to localize visual and auditory stimuli. In fact, the most prominent theory of the neural mechanisms of visual processing hypothesizes that an extensive network of cortical areas extending throughout the occipital and parietal lobes is concerned primarily with coding visuo-spatial information [67]. One reason why the spatial dimension of sensory processing is so important is that spatial representations are needed to guide movements. Thus, it is not surprising that much of the visuo-spatial processing that occurs in the `dorsal processing stream' is concerned with the transformation of visuo-spatial information into pre-motor commands relating to reaching and the control of eye movements (e.g., [19], [20], [37], [50]). Auditory spatial information also is distributed to portions of this dorsal processing stream where a similar transformation into pre-motor signals appears to occur (e.g., [13], [15], [39], [55], [69], [73]).

In humans, psychophysical evidence suggests that the neural representation of visual space is more elaborated and precise than the representation of the acoustic space. Consider for example vernier acuity, which is the finest measure of visual localization (e.g., [70], [71]). In a typical vernier acuity task, an observer is required to detect a spatial misalignment of two visual contours, as is illustrated in Fig. 1. The threshold for such judgments corresponds to retinal offsets that are substantially smaller than the diameter of individual foveal cones. Under optimal conditions, vernier acuity thresholds are approximately 5 s of visual arc (0.0014°). By this standard, auditory localization is fairly coarse. The minimum audible angle (MAA), which is the finest measure of auditory spatial offset, is slightly greater than 1° of azimuth [49]. Thus, the threshold for spatial offsets in the frontal plane is about 700-times greater for auditory stimuli than visual stimuli. Consistent with these psychophysical data are the observations that saccadic eye movements to auditory stimuli are not as accurate as visually-guided saccades (e.g., [22], [32], [35]).

The manner in which auditory position information is distributed to the oculomotor system appears to be designed specifically to permit poly-modal access to the saccadic control system [35]. Consistent with this idea is the observation that saccades to bimodal targets (auditory + visual) are faster than saccades to auditory stimuli alone or visual stimuli alone. The magnitude of facilitation cannot be accounted for by probability summation (the gain produced by two independent detection processes––one for vision and one for audition [32], [33]). As a result, inter-sensory facilitation of saccades has been interpreted as a psychophysical correlate of the convergence of auditory and visual inputs onto single neurons that is known to occur within the superior colliculus (e.g., [18], [36], [47]).

While the saccade data indicate that neural representations of static locations of auditory and visual stimuli are distributed to neural centers that generate saccadic eye movements (or reaches), it is unclear whether analogous circuits exist for auditory and visual motion. Moreover, if such circuits do exist, is there poly-modal facilitation comparable to that seen in saccadic eye movements?

To address these questions, we compared smooth pursuit eye movements (SPEM) for moving auditory and visual stimuli. SPEMs are classically associated with tracking of a moving visual stimulus. Consequently, these movements typically require a neural signal specifying target velocity (e.g., [10], [61]). We reasoned that, since SPEMs rely on motion signals, any evidence of SPEMs based on auditory motion would imply that the human auditory system extracts information specifying stimulus velocity and distributes that information to the slow pursuit control system. If SPEMs are elicited by auditory motion, we may conclude that information about auditory motion is coded somewhere within the central auditory pathways and transmitted to the smooth pursuit system.

If, however, ocular tracking of auditory motion is not possible, we are left with two possible explanations. It may be that the auditory system generates the relevant motion signals, but those signals do not gain access to the pursuit control system. Alternatively, a failure to observe SPEMs elicited by auditory motion might be attributed to an absence of motion detectors in the human auditory system. It is this latter possibility that is the primary focus of this article.

We evaluated the quality of ocular tracking in response to two types of motion: motion in the frontal plane and motion in depth. Motion in the frontal plane produces conjunctive SPEMs, whereas motion in depth produces disjunctive SPEMs, or vergence eye movements. In the case of visually-guided pursuit, it is thought that different mechanisms drive conjunctive and disjunctive SPEMs––stereoscopic cues are considered essential for disjunctive SPEMs, and motion cues are considered essential for conjunctive SPEMs. Based on the evidence that each type of movement relies on different visual cues and different motor systems (cf., [10]), we decided to study possible auditory control over each type of movement.