Scanpath Eye Movements during Visual Mental Imagery in a Simulated Hemianopia Paradigm

نویسندگان

  • Thomas G. Liman
  • Wolfgang H. Zangemeister
چکیده

10.5 (10.5) 15.75 (20.0) 10.3 (24.2) Search 6.0 (6.9) 8.5 (9.8) 7.0 (13.0) p=0.31 n.s. Task Parameter: Fixation duration in sec; median (IQR) p-value Easy Detailed Recollection Condition Viewing 0.32 (0.11) 0.28 (0.09) 0.28 (0.08) p<0.31 n.s. Imagery 0.38 (0.16) 0.35 (0.17) 0.32 (0.15) p<0.001 * comparison of picture content with all tasks together: p<0.001 ** IQR, interquartile range ++ comparison of the tasks with each other: p=0.031 n.s. +++ comparison of condition with all tasks together: p<0.001 Additional support to the “scanpath theory” are given by two studies reporting similarity between EM sequences made while viewing stimuli and those made when imagining them later (Brandt & Stark, 1997; Laeng & Teodorescu, 2002). Brandt et al. were the first to describe the similarity between spontaneous EM in visual imagery and perception (Brandt & Stark, 1997). They found that EM during visual imagery reflect the content of the visual scene. Furthermore, it has been shown that the frontal eye field analogue in monkeys showed a marked increase in metabolic activity during visual imagery (Gulyas, Persson, S., & Eriksson, 1992). Laeng and Teodorescu give further evidence that EM during visual imagery are not random but re-enact those of perception. Moreover, their findings as well as a study by Fuehr et al. (2004) demonstrate that EM are functionally involved in mental imagery processes. In their study subjects who were not allowed to perform EM during visual imagery were less accurate in subsequent memory test for visual details. Thus, eye movements during visual imagery seem to be important and play a functional role when we generate mental images. This adds further information to the growing evidence that many mechanisms including oculomotor strategies involved in perception are also used in mental imagery (Mast, Ganis, Christie& Kosslyn, 2003). EM characteristics in visual imagery Scanpath EM during visual mental imagery exhibit specific characteristics. The highly significant differences in visual imagery scanpath compared to corresponding basic viewing scanpath parameters could be summarized as a “miniaturization” or “minimization” of viewing scanpath. The different characteristics of visual imagery scanpath include smaller saccades as well as longer and less fixations. Miniaturization of scanpath in visual imagery has previously been described in healthy subjects and real homonymous hemianopia subjects (Brandt & Stark, 1997; Gbadamosi & Zangemeister, 2001). The authors concluded that different spatial and temporal characteristics of imagery scanpath reflect the underlying cognitive process of mental image generation. Thus, visualization and mental reconstruction of a visual scene or an image may not require the same temporal resolution or full-scale amplitudes for EM as in perception and correct relative spatial distances of motor activity are sufficient to support an undistorted mental image. Also, longer fixation durations suggest a longer perceptual processing time during mental visual imagery than in active viewing. The process of reconstructing the mental image seems to require prolonged fixations. Rayner and coworkers formerly described that EM parameters like fixation duration are strongly affected by the cognitive process during Journal of Eye Movement Research Liman, T. G. & Zangemeister, W. H. (2012) 5(1):2, 1-11. Scanpath Eye Movements during Visual Mental Imagery in a Simulated Hemianopia Paradigm 8 process during fixation (Rayner & McConkie, 1976). However, a recent study provides further evidence that oculomotor information encoded during visual perception is crucial for the process of proper mental image generation and they are used as spatial reference (Laeng & Teodorescu, 2002). Furthermore, a recent study could show that primary saccades are frequently followed by secondary saccades. In addition, microsaccades often occur during fixation (Ohl, Brandt & Kliegl 2011). Thus, this is a limitation of our study that has to be mentioned due to the fact that the scanpath definition neglects those two types of eye movements. Moreover, the secondary (micro-) saccades could be crucial for computation of the global/local ratio. Task and picture content effects on scanpaths Experiments by Zangemeister and Stark studied the task effect on scanpaths patterns in nonprofessional and professional subjects while viewing abstract and realistic images (Zangemeister, Sherman & Stark, 1995). They found a strong top-down component in more sophisticated viewers when looking at images. Sophisticated viewers used to favor a more global scanpath strategy. Here, we could demonstrate a major effect of oral given task on basic scanpath parameters in SH subjects, but minor influence on scanpath similarity of visual imagery and viewing. The global/local ratio as an indicator for global versus local scanning strategies, significantly differs between tasks. As we expected the task effect in the “detailed” condition elicited a more global scanning of the whole image. This effect was stronger in abstract than in realistic or search pictures reflecting in parts the visual content of the inspected image as previously described (Zangemeister, Sherman & Stark, 1995; Privitera & Zangemeister, 2007). Interestingly, we found lower scanpath similarity in the “detailed” condition. Evidently, an oral given task, such as “Look for details!” favors a more bottomup approach. However, we found less heterogeneity in similarity values concluding that the validity of string editing methods is limited for evaluation of task effects on scanpath similarity. ROI definition in string editing methods String editing is a widely accepted method for scanpath evaluation (Barton, Radcliffe, Cherkasova, Edelman, & Intriligator, 2006; Brandt & Stark, 1997; Choi et al., 1995). Our data demonstrates that both ROI definitions (geometrical “a priori” and semantic “a posteriori”) seem to be eligible for scanpath comparisons of visual imagery and perception. Throughout, we found higher similarity values and higher differences using semantic the “a posteriori” ROI definition. We concluded that the semantic ROI definition is more valid to investigate effects on top down guided gaze strategies in favor of the scanpath theory. SH and additional foveal masking In this study one further interest was the behaviour concerning scanpath EM in subjects that had to deal with a simulated homonymous hemianopia model. However, there seems to be no relevant difference in top down gaze control compared to our previous studies in homonymous hemianopia patients and normal subjects. Top down processes are of essential importance in the perception process in subjects that have to deal with homonymous hemianopia (Zangemeister, Oechsner & Stark, 1995, 1996). Besides top-down strategies, compensatory EM into the blind hemifield play a crucial role in gaze strategies of patients (Pambakian & Mannan, 2000; Zangemeister, Meienberg, Stark, & Hoyt, 1982; Zihl, 2000). Tant and coworkers as well as Zangemeister and Utz (2002) demonstrated clear parallels between simulated and real hemianopia suggesting that scanning behaviour in homonymous hemianopia is primarily visually elicited and not caused by additional brain damage (Tant & & Brouwer, 2002). A larger mask of 10° led to similar, but less efficient results, such as longer latencies to achieve and maintain fast stable saccades towards the blind hemifield (Zangemeister & Utz, 2002). Recent studies investigating simulated homonymous hemianopia support that the sensory hemianopic visual field defect is the major component of impairments of reading and visual exploration strategies found in real homonymous hemianopia patients (Schuett & Heywood, 2009a, 2009b). However, in this study we focus on scanpath comparison between visual imagery and image perception. Furthermore, we could show that an additional visual field defect with complete foveal masking has significant detrimental effects on subsequent visual imagery performance, indicated by decreased similarity between viewing and visual imagery scanpath. We conclude that missing foveal information in the preceding viewing condition deeply affects the generation of the “mental image” and therefore the consecutive top-down guided visual imagery scanpath. This conclusion puts our results in the context of the question whether visual perception and Journal of Eye Movement Research Liman, T. G. & Zangemeister, W. H. (2012) 5(1):2, 1-11. Scanpath Eye Movements during Visual Mental Imagery in a Simulated Hemianopia Paradigm 9 imagery shares identical cortical structures and common central processing mechanism. This topic and specifically representations of images in our mind have been the issue in the “imagery debate” (Kosslyn, Ganis, & Thompson, 2001; Pylyshyn, 2002). Kosslyn and coworkers propose an integrated model of visual imagery and perception. They postulated a retinotopically organized “visual buffer” in V1 that could serve as the visual working memory presenting visual information from outside during perception and from inside like mental images during visual imagery generated by higher level cortical areas (Kosslyn, 1995). They claimed that the internal image is constructed in the “visual buffer”. Patterns of activation are formed in the visual buffer either by image generating process from long-term memory or perceptual encoding process. Distance, location, and orientation of the internal image can be represented in this visual buffer and it is possible to shift attention to certain parts or aspects of it. Neural findings support that visual imagery and visual perception share common structures and mechanisms and the fact that areas of primary visual cortex preserve spatiality from the retina (Ganis, Keenan, Kosslyn, & PascualLeone, 2000; Ganis, Thompson, & Kosslyn, 2004). EM during imagery may then be connected to internal attention shifts in the visual buffer. Rayner et al. showed for the first time that missing foveal vision drastically affects reading abilities (Rayner & Bertera, 1979). Foveal masking during scene perception could demonstrate that the information for foveal stimulus identification generally is extracted during the early part of fixations (van Diepen, Ruelens, & d'Ydewalle, 1999). In their experiments an ovoid noise mask was used to replace foveal information after different preset delays following the onset of fixations. They showed that exploration is particularly disturbed when foveal masking occurs early during fixations. Considering our results, visual imagery and “filling of the visual buffer” in high level vision may be disturbed if high resolution inspection of important image features (ROIs) during preceding perception is constricted due to foveal masking. However, this hypothesis needs further validation from future EM and fMRI studies as well as further foveal masking paradigms. In conclusion, our results show that scanpath EM are involved in visual mental imagery and reflect the picture content even under SH. In contrast, additional foveal masking significantly reduces the similarity between viewing and imagery scanpath. This points toward a negative effect of foveal masking on subsequent visual imagery performance.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Is the origin of the hemianopic line bisection error purely visual? Evidence from eye movements in simulated hemianopia

It is still unclear whether the contralateral line bisection error in unilateral homonymous hemianopia is caused by the visual field defect, strategic oculomotor adaptation or by additional extrastriate brain injury. We therefore simulated hemianopia in healthy participants using a gaze-contingent display paradigm and investigated its effects on manual and ocular line bisection performance and ...

متن کامل

Are hemianopic reading and visual exploration impairments visually elicited? New insights from eye movements in simulated hemianopia.

Hemianopic reading and visual exploration impairments are well-known clinical phenomena. Yet, it is unclear whether they are primarily caused by the hemianopic visual field defect itself or by additional brain injury preventing efficient spontaneous oculomotor adaptation. To establish the extent to which these impairments are visually elicited we simulated unilateral homonymous hemianopia in he...

متن کامل

Visual imagery in hemianopic patients.

In this article we report some findings about visual imagery in patients with stable homonymous hemianopia compared to healthy control subjects. These findings were obtained by analyzing the gaze control through recording of eye movements in different phases of viewing and imagery. We used six different visual stimuli for the consecutive viewing and imagery phases. With infrared oculography, we...

متن کامل

Eye movements during visual mental imagery.

It has long been debated whether eye movements play a functional role in visual mental imagery. A recent paper by Laeng and Teodorescu presents new evidence that eye movements are stored as a spatial index that is used to arrange the component parts correctly when mental images are generated.

متن کامل

Eye movements disrupt episodic future thinking.

Remembering the past and imagining the future both rely on complex mental imagery. We considered the possibility that constructing a future scene might tap a component of mental imagery that is not as critical for remembering past scenes. Whereas visual imagery plays an important role in remembering the past, we predicted that spatial imagery plays a crucial role in imagining the future. For th...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2012