Depictive representation and pause for thought PDF

Preview

Summary

Depictive representation and pause for thought...

Description

Depictive representation and pause for thought One final point to be considered about depictive representations relates to the claim that in such representations all spatial relations are ‘immediately accessible’ (Kosslyn et al., 2003, p. 200). For instance, look at the character ‘A’ and it is immediately apparent that the shape contains an enclosed triangle, that the apex of the A is above the centre of the horizontal bar, that the apex is above the feet, etc. Now imagine an ‘A’ and, according to Kosslyn, the same visual characteristics that are immediately apparent in looking, are immediately accessible in imagining. Such a claim has an intuitive ring to it – just imagine any other capital letter and see for yourself. However, despite this cosy intuitive feeling, you might be slightly unnerved to learn that the empirical evidence is strictly against this claim. For example, findings that fit rather uncomfortably with this view come from a study by Palmer (1977). He developed something termed a mental synthesis task and this is best explained with regard to Figure 7.14. At the top of the figure are two examples of random line figures composed by interconnecting adjacent dots in a regular 3 × 3 matrix of dots. Having generated such line figures, each was subdivided into two sub-patterns or fragments where each fragment contained three lines. Now the division into the fragments was designated according to the degree to which they were considered to be good components of the original figures. As can be seen, three general categories of fragments were generated and designated as either of high, medium or low goodness. If we look on the very right-hand side of Figure 7.14, we’ll see that a triangle stands for high goodness in view of the shape being both interconnected and having a closed contour. The second shape (the larger, open triangle) has slightly less goodness since, while it has interconnected, it no longer has a closed contour. The final shape (the ‘nose’ and the horizontal line) has neither interconnectedness nor closed contours and as such stands for low goodness components.

Now on each trial in the experiment participants were presented with a pair of fragments on either side of a central fixation. Participants were told that they had to synthesise the two fragments mentally into a globally connected figure and make a response when they felt that they had completed fusing the two together. Times to make these responses were recorded. Following a 500 ms pause a complete figure was presented and the participants had to respond as to whether the figure corresponded to what the composite of the two fragments looked like. The results of the study are shown in Figure 7.15. The first thing to notice is that the time to synthesise the sub-patterns mentally varied according to their designated goodness. Responses were faster when presented with high rather than low goodness sub-patterns. These data are problematic for any simple-minded imagery account. If what the participants are doing is manipulating depictive representations of the subpatterns, then there is no reason to assume that the mental synthesis should depend on component goodness. Think about it this way. Draw one figure fragment on one overhead transparency and draw a different figure fragment on another transparency. To enact mental synthesis, simply move the transparencies together until they superimpose one another. This lateral movement in no way depends on the structural nature of what appears on the two transparencies. So if these sorts of operations were being carried out, mentally, on depictive representations, then why should the goodness effects reported by Palmer (1977) have occurred? Moreover, the decision response times in the figural matching task also showed an effect of component goodness. Responses for the high goodness cases were enacted faster than those for the low goodness cases. This is also contrary to the idea that all spatial relations...