Symbols redirect attention – in some ways, that is their intended purpose – but this “reorienting” is a surprisingly literal and involuntary effect. Even when we know symbols are irrelevant to our current circumstances, they still influence our behavior.
A simple experiment demonstrates this nicely. Hommel et al. showed that letters appearing in unpredictable locations can be identified more quickly when their location is compatible with a preceeding symbol – even if those symbols are totally irrelevant to the task, and subjects are explicitly told to ignore them!
For example, a letter appearing at the top of the screen is identified more quickly when preceeded by the word “up” or an upward-pointing arrow than when preceeded by “left” or a left pointing arrow. This is true even when the location of the target letter had been cued by more than 300 msec previously – a scenario in which people are usually slower to identify a target in a cued-location (known as inhibition of return). Again, these effects of symbols appear to be automatic and involuntary.
Maybe you’re not impressed – after all, the symbols “up” and [upward pointing arrow] definitely seem like they should be relevant to the task. But it turns out that even more abstract symbols do the same thing: you would more rapidly indicate that a circle had appeared to your right if it was preceeded by the number “8” than if preceeded by the number “2.” Conversely, if a circle on the left had been preceeded by the number “1” you’d be faster to detect it than if it had been preceeded by the number “9.” The authors of this study concluded that “mere observation of numbers obligatorily activates the spatial representations associated with number meaning.”
Why should this be so? Numbers seem to be mentally represented on a “mental number line,” in which the larger numbers on the right side of the number line correspond to regions in the right side of visual space, and vice versa for the left side. Hubbard et al’s recent NRN review article explores this symbolic “mental number line” in more detail, and reveals several other tell-tale signs of its influence on behavior.
For example, large numbers are classified as either “even” or “odd” more quickly when the response is made on the right side of space, and small numbers are classified more quickly when the response is made on the left side of space. This is true even when the hands are crossed or when responses are made with eye movements – indicating that the number-space relationship operates at a “central level” of cognitive processing (i.e., it can be elicited not only by response incompatibility but also by stimulus incompatiaibility, as when smaller numbers might be presented on the right side of space), is “effector-independent” (i.e., it can be measured both in hand and eye movements) and occurs within a “stable spatial-coordinate frame” (it is not affected by crossing of the hands). In fact, there is even a form of synesthesia in which numbers are associated with particular directions – known as number-form synesthesia.
Based on last week’s posts about the relationship of the parietal lobe to both synaesthesia and to “binding”, it will perhaps be unsurprising that the mental number line also appears to be related to parietal function. Hubbard et al. speculate that the lateral area of the intraparietal sulcus (aka, LIP) may be particularly crucial for this binding between number and space for a few reasons. First, activity in this region is negatively correlated with the distance between two numbers that are being compared in magnitude; the same area appears to be involved in a variety of “quantity comparison” tasks (including non-numerical quantities like luminance and size). Second, LIP representations seem to be “eye-centered” (compatible with the idea that the number-space relationship is maintained in a stable spatial coordinate system). Third, LIP representations seem to be “effector independent” in that it codes for spatial shifts in attention.
To what extent are these effects culturally- or developmentally-specific? Hubbard et al. review evidence that children younger than 9 do not show some of these numerical effects, and that adult Iranians seem to have a reversed mental number-line (which may relate to their writing system, which goes from right to left). In fact, the effect may even be relatively fluid within individuals: subjects will also show a reversed mental number line if they are previously presented with a clock face (in which larger numbers appear on the left side). This evidence suggests that the number-space relationship is highly dependent on learning, consistent with evidence from non-human primates in which connectivity between LIP and surrounding areas of the parietal lobe and the temporo-parietal junction (TPJ) increases after just a few weeks of exposure to tools; perhaps numbers are a kind of “symbolic tool” and allow similar relationships to develop in humans.
Such speculation aside, it is clear that the parietal lobe, and in particular the intraparietal sulcus, is active in an enormous variety of tasks. Other regions of the parietal lobe are even more mysterious, such as the precuneus – a region that is active in a huge number of fMRI experiments and sometimes thought to be involved in consciousness. Does the parietal lobe do everything?