Monday, 19 October 2009

The speed of free will

Crudely speaking, our actions can be divided into those that are automatic and driven by the environment and those that are initiated volitionally, as an act of will. In an intriguing new study, Todd Horowitz and colleagues claim to have recorded the relatively sluggish time taken for free will to be enacted. Their finding could help explain our natural tendency to search visual scenes via apparently random, haphazard attentional shifts, rather than using our conscious will to search more strategically. The fact is, our volitional control is simply too slow, rendering a deliberate, ordered approach ineffective.

In one experiment, Horowitz's team presented ten participants with a display rather like a clock face, but with capital letters in the positions where numbers would usually be. The display was visible for just a brief flash (53ms) before disappearing and re-appearing again, which it did twelve times.

In the condition that tested the speed of free will, the participants' task was to shift their gaze in time with the flashes of the display, so that they focused on each successive letter position on the clock face, starting at the 12 o'clock position and going clockwise.

The participants' aim was to look out for the letter "Y" and note its colour. Crucially, the letters changed each time the clock face returned and the "Y" only appeared during one flash of the display, in a specific position. If a participant hadn't shifted their attention around the display in time with the clock face flashes, they wouldn't be attending in the right place at the right time to see the "Y".

By varying the duration of the lulls between each flash of the clock face, the researchers were able to test the top speed at which participants were able to volitionally shift their attention from one letter position to the next. It turned out the participants needed an average of about 274ms (about quarter of a second) to make these attentional shifts successfully.

This volitional condition was contrasted with a control task in which participants could attend to any letter position on the clock face that they wanted. As before, the clock face flashed on and off and the participants had to spot the "Y" and note its colour. However, if the lulls between each flash were too quick, there wouldn't be time for the participants' automatic attentional system to shift between letter positions in search of the "Y". In this mode, with their attentional system free to operate on auto (the researchers dubbed this the "anarchy" condition), the participants needed just 85ms to shift attention from one letter position to another. Four further experiments with different parameters supported this finding.

Past research in this field has tended to use symbolic cues, such as arrows, to direct participants' volitional shifts of attention. A short-coming of these studies is that the time taken to interpret and process these cues likely contaminated estimates of the time taken to wilfully shift attention. The current research avoids this problem.

"It is substantially faster to 'delegate authority'" when searching a visual scene the researchers said. "If you tell yourself to find the letter 'P' or red verticals or your coffee mug, selective attention will shift around the visual world at a rate at least four times faster, in our estimation, than it would if you insisted on commanding each deployment of attention with an individual act of will."
_________________________________

ResearchBlogging.orgHorowitz TS, Wolfe JM, Alvarez GA, Cohen MA, & Kuzmova YI (2009). The speed of free will. Quarterly journal of experimental psychology (2006), 62 (11), 2262-88 PMID: 19255946

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.


Bookmark and Share

1 comment:

  1. When I think of clocks and discussions of free will, I think of Libet. The only thing lacking here is electrodes in the brain. Certainly, this article support the point that volition is too slow. Any further thoughts on how it connects with Libet?

    ReplyDelete

Note: only a member of this blog may post a comment.

Google+