What Your Brain Decides in a Fifth of a Second

Before you finish reading this sentence, your brain will have completed, several times over, a process science has been timing in milliseconds for more than a century and a half: perceiving something and responding.

All it takes is reacting to a light switching on. That act, so simple it barely seems “mental”, is one of the oldest subjects of experimental psychology, and one of the best at connecting the mind to the clock.

A button, a light and a stopwatch from 1868

Psychologist Arthur Jensen devotes a good part of The g Factor (1998) to a kind of experiment that seems too simple to matter: reaction time.

We owe the method to Dutch physiologist Franciscus Donders, who in 1868 designed an ingenious way of measuring mental processes that can’t be observed directly:

• In simple reaction time, the person holds down a key and releases it as soon as a stimulus appears (a light, a sound). There’s nothing to choose, just react.
• In discrimination reaction time, one of two possible stimuli appears (a yellow or a blue light, for example), and only one of them calls for a response.
• In choice reaction time, each stimulus has its own response: yellow light, left button; blue light, right button.

Donders’s idea, the subtraction method, was simple: subtract the time for the simple task from the time for the complex one, and what remains is what the brain needed for the extra step (discriminating, choosing). For the first time, a mental operation had a measurable duration.

A fifth of a second, and what gets added on top

In young adults, simple reaction time averages around 200 milliseconds: a fifth of a second.

From there, each layer of processing adds something. According to the data Jensen compiles, discriminating between two stimuli adds between 30 and 100 milliseconds, depending on how hard they are to tell apart, and choosing between different responses adds more time on top of that.

Hick’s law: every doubling costs the same

In 1952, British psychologist William Hick described precisely how that extra time grows. As the number of possible alternatives increases, choice time doesn’t grow in a straight line but roughly logarithmically: going from two to four options adds about the same as going from four to eight.

Hick interpreted this through Claude Shannon’s newly born information theory: each doubling of the alternatives amounts to one more “bit” of information to process, and the brain seems to take a fairly constant amount of time per bit. Ray Hyman confirmed the regularity in 1953, and today it’s known as the Hick-Hyman law. It’s even used in interface design, to explain why menus with too many options slow decisions down.

Jensen built on that law the apparatus that bears his name, the “Jensen box”: a console with a central button and eight lights around it, of which 1, 2, 4 or 8 would light up. That way he could measure both each person’s reaction speed and how much they slowed down as bits of information were added.

What reaction time says about intelligence

This is where the finding that links all of this to the g factor comes in: that time, so short it’s measured in milliseconds, correlates with IQ, and the correlation tends to grow with the complexity of the task. In the data Jensen summarizes, it goes from about -0.10 for simple reaction to -0.30 for choice reaction (negative because less time goes with higher IQ).

The problem with many of those studies was the sample: university students, with a narrow range of ability that tends to shrink correlations. That’s why the work published in 2001 by Ian Deary, Geoff Der and Graeme Ford matters. They tested 900 people aged 56, representative of the Scottish population, with an intelligence test and an apparatus for simple and four-choice reaction time:

• The correlation with simple reaction time was -0.31.
• With four-choice reaction time it rose to -0.49.
• The correlations were similar when broken down by sex, social class and education level.

These figures are considerably larger than those from the classic studies with university students. In a review of 172 studies published over five decades, Sheppard and Vernon (2008) concluded that the link between processing speed and intelligence is modest but highly consistent: it shows up across very different tasks, ages and populations.

Why the simplest task says the least

There’s a nuance that often goes unnoticed: the simpler the task, the less it says about differences between people.

In simple reaction time, factors unrelated to cognition weigh heavily, such as momentary attention or muscle speed. In choice reaction there are more mental steps involved (perceiving, discriminating, deciding, executing), and therefore more room for processing speed to show.

Jensen sums up this idea with a thesis that runs through his book: the “active ingredient” in these correlations isn’t muscle speed or sensory acuity, but g, the general factor that’s also at work in far more complex tasks, such as solving an abstract reasoning problem. Other researchers read it more broadly, as an indicator of the overall efficiency of the nervous system; Deary’s own group later found, in that same cohort, that slower reaction time was also linked to higher mortality (Deary and Der, 2005), which points to something more than intelligence in the strict sense.

What it does measure, and what it doesn’t

This finding supports some conclusions and not others:

• It does show that even the most elementary cognitive tasks, measured with millisecond precision, bear a statistical relationship to general intelligence.
• It doesn’t mean that a person who is slow to react to a light has less intellectual ability: even a correlation of -0.49 leaves a great many exceptions, and reaction time varies with fatigue, age, attention and practice.
• It isn’t a diagnostic test and doesn’t replace any serious psychometric assessment.

What this century-old experiment leaves us with is a striking idea: you don’t need a complex problem for differences between people to show up. A light, a button and a stopwatch that can measure milliseconds already record something about how each brain processes information.

If you’re curious how the brain itself reacts, with no conscious response involved at all, to an equally simple stimulus, your brain responds to a click before you know what it is explains the equivalent experiment using brain waves.

Sources

• Deary, I. J., Der, G. and Ford, G. (2001). Reaction times and intelligence differences: A population-based cohort study. Intelligence, 29(5), 389-399. DOI
• Deary, I. J. and Der, G. (2005). Reaction time explains IQ’s association with death. Psychological Science, 16(1), 64-69.
• Donders, F. C. (1969). On the speed of mental processes (trans. W. G. Koster; original 1868). Acta Psychologica, 30, 412-431.
• Hick, W. E. (1952). On the rate of gain of information. Quarterly Journal of Experimental Psychology, 4(1), 11-26. DOI
• Hyman, R. (1953). Stimulus information as a determinant of reaction time. Journal of Experimental Psychology, 45(3), 188-196.
• Jensen, A. R. (1998). The g Factor: The Science of Mental Ability. Praeger.
• Sheppard, L. D. and Vernon, P. A. (2008). Intelligence and speed of information-processing: A review of 50 years of research. Personality and Individual Differences, 44(3), 535-551. DOI