Moving the eyes but not looking - why do we do it?

You've probably noticed how people move their eyes about when in the midst of conversation, often in ways that have absolutely nothing to do with looking at the scene around them. In fact these 'non-visual gaze patterns' also occur when we're on our own, in the complete dark, and even when our eyes are closed. The implication is these eye movements are a result of mental processes that have nothing to do with vision or social factors.

To investigate, Dragana Micic and colleagues recorded the eye movements of 17 female and 12 male participants while they performed two tasks designed to be as similar as possible except for the fact that one required delving into long-term memory and the other didn't. The testing took place in a bare room with the walls draped in white sheets to reduce visual stimulation.

The first task involved the participants sitting on their own in the room, listening to three words then repeating them back after a delay (drawing on short-term or working memory). The second task was the 'remote associates test', which involves delving into long-term memory to identify the one word that matches the meaning of three others (e.g. envy, golf, beans; answer: green). All words were presented out loud, so there was no reading. The participants made far more jerky 'saccadic' eye movements during the latter task, thus suggesting that non-visual eye movements are triggered by long-term memory retrieval.

Two further studies tested whether these eye movements occur more when digging for a less accessible memory and secondly, whether they actually serve a useful function. This was achieved by varying how well the participants learned a word list (thus varying the accessibility of the memories) and, in another experiment, by instructing participants to fixate during a long-term memory retrieval task. These tests revealed that eye movements don't increase when attempting to locate a particularly inaccessible memory, and neither do they play a functional role - participants' memory performance was not impaired when they were instructed to fixate. In the jargon, this suggests that non-visual eye-movements are an epiphenomenon - triggered by long-term memory retrieval but playing no useful part in that activity.

So, why should accessing long-term memory trigger eye movements? Micic's team think the phenomenon is caused by an evolutionary hang-over. By this account, some of the same (evolutionarily older) processes involved in searching a visual scene are co-opted for use when consciously exploring the mind's archives. 'The possibility that spontaneous saccadic eye movements occur despite being non-functional may simply indicate how the addition of higher brain functions does not necessitate the "redesign of the whole brain",' they said.

At first, these new findings seem to contradict another line of research that's previously shown the beneficial role of gaze aversion (looking away from a questioner) when children and adults are working out a solution to maths and memory problems [pdf]. And to contradict the literature showing that wiggling the eyes from side to side can aid memory. Micic thinks the difference here is probably between involuntary non-visual eye movements triggered by memory processes (the focus of the current study) and voluntary eye movements, as in the eye wiggling and looking away.
_________________________________

Micic D, Ehrlichman H, and Chen R (2010). Why do we move our eyes while trying to remember? The relationship between non-visual gaze patterns and memory. Brain and cognition, 74 (3), 210-24 PMID: 20864240