Educación

Our findings showed a global also for hand movements, even in the single-task condition in the absence of eye movements. Thus, the global effect is not specific to the oculomotor system, but also occurs in the limb motor system. Moreover, the global effect was mostly similar for eye and hand movements. It is known that the global effect occurs on the level of target selection, i.e., before the process of separating the target from the distracter has been completed. Thus, eye and hand are coupled at this early stage. To achieve this coupling, eye and hand at least have to exchange information, or, even more, share one target representation. However, it is unclear, which alternative may be the case. Importantly, the term “shared target representation” as used here also implies an identical point in time when information about the target is read out. If either the target representation itself or the point in time when it is accessed are different for eye and hand, we speak of separate target representations.

In experiment 1 and 2 the global effect of hand movements was of the same direction and magnitude as that of eye movements. This similarity requires at least an exchange of information. The fact that with increasing eye latencies the global effect was reduced for the eye, but not the hand, raises doubt on the idea of a shared target representation. If the distracter affected a shared target representation, then the dependence of the global effect on latency should be the same for eye and hand. Thus, this finding speaks for two separate target representations. For the hand, the absence dependence of the global effect on latency suggests that the target representation is not updated, but “frozen” on an early stage.

However, this difference in the latency dependence of the global effect between eye and hand was not observed in experiments 2 and 4, because the global effect for the eye was no longer latency dependent. This seems unusual compared to the findings of other authors (e.g. Findlay, 1982; Ottes, van Gisbergen, & Eggermont, 1985). It may be due to the response in the present experiment being typically executed late in the target selection process, i.e., when separation of the target from the distracter is nearly finished. This explanation is supported by

the finding that in all experiments the global effect was not very large, i.e., movements did not land exactly in-between the target and the distracter, but closer to the target. Thus, the fact that eye and hand showed a similar behaviour in experiments 2 and 4 with regard to latency dependence does not necessarily speak for a shared target representation.

In experiments 1, 2 and 4, the dependence of amplitude on latency – irrespective of the presence of a distracter - differed for eye and hand movements. Hand movement amplitude always increased with latency, and eye movement amplitude first increased before again decreasing. This difference could be induced by variations in processes after target selection, like the continuous integration of information about the target and the actual status of the effector (eye or hand). Therefore, this result cannot be interpreted as evidence for or against separate target representations of eye and hand.

Summing up, the results of experiment 1 and 2 represent only weak evidence for separate target representations. With truly separate target representations, the global effect for eye and hand should be different under certain conditions. Such a difference was observed in experiment 3, where the global effect was in the same direction but larger for eye than for hand movements. A global effect different in magnitude for eye and hand can not be explained by the distracter acting on a shared target representation.

The strongest argument against the hypothesis of shared target representations results from the different behaviour of eye and hand towards the “near” distracter in experiment 4. Whereas the global effect for the eye was the same as in experiments 1, 2 and 3, there was no global effect for the hand with a distracter less eccentric than the target. Hence, the global effect is less stable for hand movements than for eye movements, but seems to depend more on the features of the target-distracter configuration. This difference between eye and hand speaks against a shared target representation.

The remaining question is why there was a global effect for the hand in experiment 1 with the near distracter, but not in experiment 4. One major difference between experiment 1 and 4 was the degree of predictability of target position. Whereas in experiment 1 the distracter was always more eccentric than the target and in experiment 2 it was always less eccentric, in experiment 4 it could occur either more or less eccentric than the target. To explain the different behaviour of eye and hand towards the near distracter in experiment 4, the concepts of distracter saliency and inhibition have to be introduced.

According to Tipper and colleagues (e.g., Tipper, Howard, & Jackson, 1997; Tipper, Howard, & Houghton, 2000), selection is performed by inhibiting the activation caused by the distracter. With the right amount of inhibition, eye and hand land exactly on the target. With too little or too much inhibition, they do not, i.e., an error occurs. With too little inhibition, the

error is in the direction of the distracter. Movements land in-between the target and the distracter, as found both for the eye (global effect) and the hand (e.g., Tipper et al. 2000; Welsh, Elliott, & Weeks, 1999). With too much inhibition, the error is in the direction opposite to the distracter, i.e., eye (e.g., Doyle & Walker, 2001; Sheliga, Riggio, & Rizzolatti, 1995) and hand movements (Fischer & Adam, 2001; Tipper et al., 1997) veer away from the distracter. The amount of inhibition is believed to increase with the salience of the distracter. Salience may be expressed in properties of brightness or contrast, and for hand movements also in distance from the hand (Tipper et al., 1997). Thus, the near distracter generally has differential salience for eye and hand. Assuming that the near distracter was inhibited more strongly for hand movements, the global effect persisted for eye movements, but disappeared for hand movements.

However, the same should apply to experiment 1, as there was also a near distracter involved. In contrast, a global effect for the hand was observed. In experiment 1 subjects knew that the distracter was always the nearer stimulus, so the direction in which the would occur was known in advance. We can only speculate that this higher degree of predictability reduced the saliency of the distracter and therefore, also the amount of inhibition necessary to act on the distracter. This would account for the observed global effect for the hand. In addition to the differences in predictability, experiments 1 and 4 differed also with regard to the features defining the target, the spatial distribution of distracter positions, and the complexity of visual processing required to select the target. The differences in hand movement responses towards the near distracter in experiments 1 and 4 could also be due to any of these factors.

On the basis of the results discussed so far, it seems unlikely that the target representation affected by the distracter is the same for eye and hand. Instead, we assume that parallel processes of response selection and inhibition influence the separate target representations of eye and hand. However, an interaction in the sense of an exchange of information between these processes seems to be necessary to achieve a global effect as similar for eye and hand as in experiment 1 and 2. These somewhat mixed findings may be explained by the amount of information exchanged being dependent on the conditions of the task. The difference in the global effect for eye and hand in experiment 4 with the near distracter suggests that in this case, less information is exchanged.

One possibility to gain further insight into this exchange of information is the application of the dual-task methodology. Although comparing the single- with the dual-task condition irrespective of the global effect does not provide the advantage of specifying the stage at which coupling takes place, the results are worth a closer look.

3.7.2 Interaction of eye and hand in the dual-task – spatial and temporal coupling