Cognitive modulation of midbrain function: task-induced reduction of the pupillary light reflex

Abstract

The activation of processing resources has widespread effects in the nervous system. A model of pupillary control systems (Steinhauer S.R., Hakerem G., 1992. The pupillary response in cognitive psychophysiology and schizophrenia. Ann. N.Y. Acad. Sci. 658, 182–204) had predicted that ongoing cognitive activation should result in inhibition of the light reaction at the level of the oculomotor nucleus, n. III, in the midbrain. In this study, modification of parameters of the pupillary reaction to light were examined during varying task demands. The averaged light reaction was recorded from 33 male and female healthy volunteers during the performance of a serial 7 subtraction task and compared to a ‘no task’ condition. For 13 subjects, an additional verbalization task with little processing demand (add 1) also was presented. Two types of effects were observed. Firstly, the tonic pupil diameter increased from the no task to the easy (add 1) task, and increased further in the more demanding condition (subtract 7). Secondly, the extent of the phasic light reaction was significantly reduced and the latency at the end of the contraction was significantly decreased in the ‘subtract 7’ condition compared to both the no task and easy conditions (which did not differ from each other). The locus of interference with the light reaction was the Edinger–Westphal complex of the oculomotor nucleus, which is the motor center for the pupillary sphincter muscles. Descending cortical influences inhibited the activity of the Edinger-Westphal complex. Thus, increasing activation had a tonic inhibitory effect on this center, while higher levels of processing complexity produced a separate component of inhibition that interacted with dynamic activation at this midbrain site. It was suggested that the variation in the light reaction is quantitatively responsive to varying processing loads, and may be utilized as a sensitive metric for a wide variety of cognitive operations.

Introduction

Assessing the effects of varying cognitive demands on the nervous system has been one of the most enduring problems posed to cognitive psychology in the past three decades. One suggestion early in the development of processing evaluations was the measurement of changes in the pupil diameter as a sensitive indicator of resource allocation and processing load (e.g. Kahneman, 1973, Janisse, 1977, Beatty, 1982).

Among other approaches employed for assessing the relative demands engendered by different cognitive operations is the use of dual-system measurements. Activation associated with a particular task is inferred, for example, by the attenuation of EEG alpha power (Davidson et al., 1990), or the reduction of several components of the event-related potential to a separate task (Kramer et al., 1995). One difficulty in the evaluation of changes that impinges on these measures is that they are determined by a multiplicity of neural sources. The present paper suggests an adjunctive procedure for similar evaluations in which a more limited and better understood system contributes to the primary response — the pupillary constriction to light. The pathway for the pupillary light reaction provides an optimal model for investigating sensory and psychological inhibitory influences associated with varying modes of activation.

The reduction in the extent of the light reaction during psychosensory stimulation in animals and humans has been well established (Lowenstein and Loewenfeld, 1963, Loewenfeld, 1993). In humans, Hess (1975) reported that the light reaction decreased in amplitude as the rate of auditory clicks was increased. Frith (1976) reported an increased diameter and decreased light reflex amplitude when the subjects were exposed to a 95-dB tone. Inhibition of the pupillary light reflex was also demonstrated in anticipation of fear-provoking stimuli (Loewenfeld, 1993, Bitsios et al., 1996), and by the initiation of a simple motor response (Gavriysky, 1991).

Arousal associated with anxiety contributes to inhibition of the light reaction. Patients with anxiety disorders have shown decreased light reactions compared to control subjects, though initial diameters do not differ (Bakes et al., 1990). Decreased light reflexes were observed in normal subjects in whom fear was induced by a threat (but not actual delivery) of shock (Bitsios et al., 1996, Bitsios et al., 1998).

The afferent pathway from the retinal receptors (including the crossed and uncrossed optic nerve and optic tract inputs) involves only a synapse in the pretectum before secondary crossed and uncrossed projections reach the motor center in the third nerve (oculomotor) nucleus. The efferent pathway involves a synapse at the ciliary ganglion, with the final motor neurons synapsing on the pupillary sphincter to provide the major component of pupillary constriction. In the studies cited above, the stimulation of arousal systems, especially those associated with the reticular activating system, was most strongly implicated. The electrical stimulation of peripheral nerves of ponto-mesencephalic reticular formation has been shown to elicit pupillary dilation, electroencephalographic arousal, and inhibition of parasympathetic activity (Bonvallet and Zbrozyna, 1963).

Additional inhibitory influences on the parasympathetic center for pupillary constriction, the Edinger–Westphal region of the oculomotor nucleus, include descending cortical afferents (Loewenfeld, 1993). Thus, it is likely that cortical influences, including cognitive activation, would have an effect on pupillary reactions to light. There are minimal findings related to the cognitive activation and inhibition of the light reaction. In a prediction task, in which either brief (1 ms) visual or auditory stimuli could be presented, the light reaction was decreased in amplitude by approximately 10% as compared to a condition of certainty (Steinhauer et al., 1979). In that situation, each stimulus eliciting the light reaction was also conveying information regarding specific feedback, which leads to pupillary dilation (Friedman et al., 1973). The short duration of the stimulus also precluded the evaluation of secondary constriction and redilation components of the light response that are present only when longer duration stimuli (e.g. 1 s) are employed (Loewenfeld, 1993).

What has not been established is whether ongoing cognitive processing unrelated to the stimulus could influence the light reaction. In developing a model to account for varying patterns of pupillary motility during different cognitive operations, it was hypothesized that maintained cognitive activity should have an inhibitory effect on the parasympathetic pathway of the pupil (Steinhauer and Hakerem, 1992). The present study aimed to address this question by invoking a task that would maintain an ongoing cognitive load at the same time that pupillary reactions to light, unrelated to the task, were recorded. Probable modifications of the pupil diameter would be expected to produce tonic effects on the diameter (increased sympathetic or reduced parasympathetic tone, leading to an overall increase in diameter), as well as the phasic reduction of constriction to light (reduced amplitude of constriction due to inhibition of parasympathetic activation).