Oculomotor behaviour in healthy subjects, stroke patients and a case of visual form agnosia.
PhD thesis, University of Glasgow.
Full text available as:
In my thesis I mainly focus on Milner and Goodale’s model (1995, 2006, 2008) of two visual pathways. While the dorsal stream is supposed to be involved in on-line action, i.e. to deal with the immediate and accurate response to a present target, they state that the ventral stream comes to action when previously perceived and memorised visual target characteristics are required for memory-guided action (off-line action).
A lot of evidence for the existence of these separate pathways has come from visual form agnosia patient DF who has repeatedly shown an impaired performance for off-line tasks while she has repeatedly shown an almost flawless performance on on-line tasks (e.g. Goodale et al., 1994a). In DF, this functional dissociation is supposed to be corroborated by her relatively spared dorsal and impaired ventral streams respectively (James et al., 2003).
Likewise patients with hemispatial neglect show a pattern similar to patient DF with off-line reaching impairments such as deficits in anti-pointing and delay tasks and relatively spared on-line actions (Rossit et al., 2009b, 2011). Indeed, hemispatial neglect occurs frequently after lesions to the right inferior parietal lobe (e.g. Mort et al., 2003) and Milner and Goodale (1995) speculate that the IPL gets input from ventral stream regions, which would explain the observed deficits in off-line actions. However, due to the heterogeneity of the lesions in patients with hemispatial neglect, an anatomical argument is much more difficult to make.
In this thesis I firstly aimed to examine the oculomotor behaviour of neglect patients and secondly of visual form agnosia patient DF in a series of experiments that tap into either on-line or off-line eye-movement tasks to establish whether Milner and colleagues’ (Milner & Harvey, 2006) action dichotomy can be upheld for the oculomotor domain.
In the first experiment I aimed to find an answer to the question of whether the bilateral anti-saccade impairment (Butler et al., 2009) is the result of a vector inversion deficit (inability to perform off-line actions) or an inhibition problem. To do that I expanded Butler et al.’s study (2009) on pro- and anti-saccade tasks by testing the patients’ ability to inhibit saccades in an additional fixation condition. In line with Butler et al.’s (2009) study my neglect patients executed many erroneous pro-saccades in the anti-saccade condition and they also showed neglect typical leftward biases in the pro-saccade condition. Furthermore, the results showed that most of the neglect patients were able to withhold eye movements towards targets. As they did not show a general severe inhibition problem it is very likely that the erroneous pro-saccades in the anti-saccade task were caused by a deficit to perform off-line actions rather than by an inhibition problem.
These findings were further corroborated in experiment 2 in which neglect patients were asked to perform a more complex fixation task with interleaved fixation and pro-saccade trials. Although the patients performed worse than the controls, they were able to withhold most eye-movements during the fixation trials. Thus the occasionally executed erroneous pro-saccades in fixation trials might reflect the greater demands of the complex fixation task rather than a general inhibition problem.
The third experiment examined immediate, stimulus-driven (on-line) and delayed, memory-guided (off-line) saccades. The results showed that all patients were more impaired for the off-line saccades than for on-line action. However this impairment might not be neglect specific as no difference was found between stroke patients with and without neglect.
The fourth experiment focused on the ability to perform oculomotor on-line corrections towards perturbed targets that could suddenly and unpredictably change in location. This task required the on-line adjustment of eye-movements to follow the target. Most of the neglect patients were able to correct their saccades in these perturbed trials and general impairments were often connected to parietal lobe lesions, which might involve the visual dorsal stream.
Experiment 1, 3 and 4 were also carried out on patient DF. She showed no general problems in performance in the pro-saccade (on-line) and fixation condition in experiment 1, yet she was impaired on anti-saccades (off-line). In experiment 3 she was able to execute saccades towards presented lines but was again impaired in the off-line condition (delayed lines). In experiment 4 she showed no problems to perform on-line corrections towards perturbed stimuli.
In summary, on a functional level my results support the distinction between on- and off-line tasks that has been established through the use of pointing and grasping tasks, which I have now extended to the oculomotor domain. The neglect patients, as well as patient DF, were impaired for the tested off-line actions while they showed no general deficits for on-line actions.
Actions (login required)