Attentional demands of continuously monitoring orientation using vestibular information

Abstract

The aim of this series of experiments was to determine whether attention is normally required for continuously processing vestibular information concerning orientation, or is required only when orientation is disrupted (eg by vestibular dysfunction or by conflicting visual and vestibular orientation cues). In the first two studies, healthy subjects were passively oscillated, and indicated when they perceived they were passing through their starting position. There was only weak evidence for interference between performance on this ‘continuous orientation monitoring task’ and on concurrent mental tasks. However, a third study showed that when patients with vestibular imbalance carried out the continuous orientation monitoring task their performance on a concurrent mental arithmetic task was substantially impaired. This dual task interference was correlated with inaccuracy in judging orientation on the continuous orientation monitoring task, which in turn correlated with severity of recent vestibular symptomatology (assessed by questionnaire). In a fourth experiment, disorientation was induced in healthy subjects by rotating the visual field about the line of sight. Bidirectional interference was observed between monitoring orientation (assessed by accuracy in setting a rod to the perceived vertical) and performance of an arithmetic task. Dual task interference was correlated with baseline levels of disorientation induced by the visual field, as indicated by inaccuracy in judging the visual vertical. These findings suggest that monitoring orientation makes significant demands upon cortical processing resources when disorientation is induced, whether the disorientation results from deficient sensory functioning or from ambiguous perceptual information.

Introduction

In order to coordinate our physical activity and postural control with respect to the environment, we continuously monitor and update our position and motion in space by means of central integration of information from the visual, somatosensory and vestibular senses. However, it has been shown that vestibular information alone can be utilised to update representations of position in space, which have been assessed by passively rotating vision- deprived subjects and then requiring them either to report their position verbally [5], [19], [26], to reposition themselves to their original orientation [10], [17], [31], [36], or to orient their eyes towards the remembered location of a visual target presented before they were moved [4], [18].

The cortical structures that are involved in the central processing of vestibular information for orientation and navigation have been investigated by means of neuroimaging [7], [23], [35] and studies of patients with cortical lesions [16], [20], [33]. While it seems evident that the parieto-insular cortex plays a crucial role in spatial processing, including the processing of vestibular information [8], additional areas are known to make an important contribution to spatial representation. These include the hippocampus [14], [35], accessed from the parieto-insular cortex or from subcortical structures, as well as head direction cell systems [3].

Spatial updating has been characterised as an ‘automatic’ process, in the sense that it occurs involuntarily, and cannot be suppressed without mental effort, when forming a spatial representation [30]. However, it may not be ‘automatic’ in the sense of making no demands on attentional capacity [30], [24]. The dual task technique provides a means to examine the extent to which different activities demand attentional capacity, ie draw on common central processing resources. In dual task experiments, decrement in performance occurring when two tasks are performed simultaneously (relative to baseline performance on each single task) indicates that both tasks demand attention. The nature of that common resource might be general, for example, the two tasks might depend upon ‘central executive’ control which is needed for coordinating multiple task performance [2]. Alternatively, the resource the two tasks share may be more specific; for example, if both tasks involve spatial processing then they might compete for specific spatial processing resources [25]. The advantage of using dual task experimentation in healthy subjects to complement studies of patients with cortical lesions is that adaptation following injury may result in idiosyncratic alterations in central processing.

There have been remarkably few dual task studies of the attentional processes involved in monitoring orientation using vestibular information. However, it has been shown that counting backwards can interfere with orientation when walking without vision [6], [32], a task which involves the use of both vestibular and non-vestibular information [28]. Moreover, Yardley and colleagues have demonstrated interference between healthy subjects’ performance on tasks involving purely vestibular perception of orientation and mental tasks (ie reaction time to an auditory stimulus, counting backwards) [39], [40], [42].

The results of Yardley and colleagues were interpreted as indicating that attention is required for vestibular perception of orientation; ie vestibular information depends upon central mental processing, and suffers from capacity limits. However, in all of these studies subjects were required to indicate their orientation after being passively moved. The advantage of this methodology was that the mental task was performed during self-motion but the orientation response was carried out after the mental task and self-motion had been completed; thus interference could not have simply resulted from concurrent performance of two responses (ie responses to the orientation and mental tasks). But in order to perform such a task the participant does not simply process sensory orientation information—in order to utilise this information to generate an accurate response after the self-motion task is completed, the information must also be stored as a representation of orientation in memory. Therefore it remains possible that attention was only required in order to encode or maintain a spatial representation in memory, but not for processing vestibular information online.

In order to examine whether attention is required for online processing of vestibular information concerning orientation, a new task was devised. For our new ‘continuous orientation monitoring’ task, subjects were continuously rotated in a motorised chair through a pseudo-random series of brief turns of varying amplitudes and alternating direction. Subjects were required to press a button whenever they judged themselves to be facing a reference point at a fixed location within the room (ie their initial starting position or ‘origin’), and therefore had to continuously update their perceived spatial position relative to the origin. To establish the accuracy of continuously monitoring orientation using only vestibular information, an initial pilot experiment was carried out to compare responses in the light (when visual information could be used) and in darkness (when only vestibular information was available).

A series of experiments in healthy subjects and patients were then used to investigate whether continuously monitoring orientation normally demands attention, or whether it demands attention only when deficits in monitoring are induced, either by vestibular dysfunction, or by experimental manipulation of perceptual conditions. If processing sensory information in order to continuously monitor orientation can normally be accomplished with little or no requirement for attention, healthy subjects’ performance on the continuous orientation monitoring task should be unaffected by a requirement to simultaneously carry out a mental task. In the first two studies we therefore compared healthy subjects’ accuracy in monitoring orientation with and without the requirement to carry out a concurrent mental task. To test the hypothesis that monitoring orientation does require attention when orientation is disrupted by sensory deficits or ambiguous perceptual information, the third study compared the single and dual task performance of patients with unilateral vestibular dysfunction, and determined the relationship between symptoms and task performance, while the final study examined the effect of experimentally induced disorientation on concurrent performance of an orientation and an arithmetic task in healthy subjects.