Eye movement patterns during imagery are determined by the eye movement patterns during perception

Visualization without eye movements is the only visual meditation technique conducive to Samadhy

Dinu-Stefan Teodorescu

Tromsø University

Three different perception styles, Free Vision, Fixation and Unfocused or Holistic Vision, have been investigated in twelve participants, divided in three experimental groups by recording their eye movements while they viewed and visualized eight irregularly-checkered diagrams. The same eye movement patterns were observed during viewing the stimulus and during the imaging of the stimulus. Vividness and accuracy tests were used to quantify the differences between the three perception styles. Contrary to previous findings (Neisser, 1967), the most vivid imagery was found in the Holistic Visual group. Greater accuracy in recall was found in the Free Vision group. The flexibility of the attention window was hypothesized to account for the different perception styles and the findings propose an enlargement of the imagery theory of Kosslyn (1994). Strong evidence that scanpaths during perception are encoded in a kinesthethic memory was found from the repetition of the same eye movement patterns during retrieval in the visual imagery condition. It is proposed that only the Holistic Vision and Fixation styles in perception are conductive to an absence of eye movements, stabilization of image and achieving Samadhy during meditative visualization.

INTRODUCTION

Key words: holistic perception, attention window, stabilization of image.

Visualization of sacred objects, like mandalas (Kværne, 1995; Tucci, 1969), yantras (Rao, 1988; Khanna, 1997; Bru-sgom rGyal-ba g.yung-drung, 1996; Johari, 1987), icons ( Laeng, 1980; Dancu, 1982), and kasinas ( Buddhaghossa, 1975), have been kept in high esteem from the world mystical traditions, used for attaining of one or another types of Samadhy ( Patanjali, 1990). The metaphor that the eyes are the window of the soul have been taken literally by transpersonal researchers in the studies of eye movements, all trying to decode the unseen Kognitive ( Wilber, 1999) processes. Vision has certainly a privileged place in the human cognition, since most areas of the brain appear to be involved in visual processing. Indeed; it is since Aristotle that thought is supposed to be also visual, and after the apparition of transpersonal psychology, visualization have been enthroned again as an important area of research. The present research have been done in the spirit of the new integral psychology ( Wilber,in press) using the tools from the third quadrant to account for changes in the first quadrant ( Wilber, 1995), trying to establish a new tradition for research in the Transpersonal.

It is possible and desirable to use all the tools we have in order to validate the information of different quadrants, so that the Transpersonal science may become really holistic, an all quadrant research.

The present study was done in order to clarify the subject of visualization techniques that have a broad following, especially in the indian tantric traditions (Feurstein, 1998; Johari, 1986 ) , tibetan tantric traditions ( Tucci,1967; Kværne, 1995) and to add to the researched data of Transpersonal Psychology.

Visual perception have been taken more or less to be of only one kind, namely the normal scanning of the visual field, following a “ hypothesis testing” strategy (Gregory, 1998), when the information is extracted during the fixations and integrated during the scanpaths ( Neisser, 1967; Hebb; 1967; Stark, 1971 ). The visual perception styles have been not so much researched, but the practical implications of “special “ perceptual styles has probably always been appreciated in human areas such as sports, e.g. basket, martial arts ( Herigell, 1974; Deshimaru, 1983 ) and mysticism (Bankei,1984; Buddhaghosa, 1970 ). The visualization practices used for meditation purposes, has as its main point the stillness of mind using an image, first using visual perception and afterwards imagery. The apparition of deeper brain waves , such as alpha, theta and delta are used to indicate the achievement of a meditative state (Motoyama, 1978; Hirai, 1994; Tart, 1980; Shapiro & Walsh, 1985 ).

Eye movements are associated with beta waves, and hence a meditative state is excluded, which is associated with alpha, theta and delta brain waves. It is therefore claimed that there is a clear difference between a normal visual perception and a holistic visual perception, which in essence could be captured by the difference between a serial cognitive processing ( as in the normal scanning visual perception ) and a parallel cognitive processing (supposedly used in holistic visual perception ). Although we cannot measure directly these different mental processes, by using eye movements as the target of analysis, we might infer correctly the presence of these different processes during visual cognition and imagery. Several studies have acknowledged vaguely the implication for a parallel visual processing, for example, that vivid imager can perceive literally ( Sheehan, 1967 ), eidetic imagers have a different encoding strategy ( Luria, 1978; Haber & Haber, 1964 ) and Norton and Stark ( !971 ) noticed that in recognition studies, the participants seemed to recognize the image even before they scan the whole image. Paivio ( 1986, 1991 ) in his dual-code theory, has proposed the simultaneous encoding of pictures in the visual code, as opposed to verbal information that uses a sequential processing ( Goldenberg, 1987 ), this clearly pointing out a parallel processing system.

The perceptual learning theory ( Hebb, 1968; Hochberg, 1970 ) proposed that thoroughly scanning which brings parts of a stimulus serially into the foveatic vision is the only way to effectively encode visual stimuli. All other kind of vision, for example peripheral visual perception was able only to distinguish vague, gross features such as brightness or contour. However, several studies have shown that the peropheral vision can be quite accurate ( Maurer & Lewis; 1981, Cohen, 1981 ) although one of its roles is to guide the foveal perception ( Irwin, 1992; Finke, 1985, Rayner et al.,1978; Yarbus, 1967 ) towards the “interesting” targets ( Gibson, 1998 ).

The representation in the brain of visual information while scanning an image have been proposed to be encoded in two separate representational systems: one encoding the visual information and the other encoding the kinesthetic information ( Hochberg, 1970; Kosslyn, 1990; Gauthier et al., 1990; Henderson, 1994; Schneider, 1995; Demarais & Cohen, 1998 )

Scanpath and visual attention theories

The prevailing paradigm in eye movement research (Rayner et al.,1978) is the scannpaths theory which proposes that visual information is integrated across fixations of the eyes. This theory was also formally proposed by Norton and Stark (1971), based on the research of Yarbus (1967) as well as some theoretical insights of Hebb (1968). The theory proposed a «feature ring» , later expanded into a «feature network» which is an integration of sensory and motor elements which guide the saccades of the eye for each visual image. The feature network is supposed to use serial encoding, and at retrieval during a recognition task, the same scannpaths is supposed to be reenacted. The theory was acknowledged to include many problems even from the beginning (Norton & Stark,1971). In later revisions (Stark & Ellis, 1981; Brandt & Stark, 1997), the theory is still wanting because of the dismissal of the role of peripheral vision in the generation of the scanpaths (Brandt & Stark,1997). One proposal is that there is a transsaccadic memory that is limited up to four tokens or object files (Irwin,1998), a memory type close to short -term visual memory. The transsaccadic memory theory have been further developed and renamed as «object-file» theory of transsaccadic memory (Irwin,1996), using the theoretical framework for object perception proposed by Kahneman and Treisman (1984). Specifically, Irwin (1998) used for the development of his theory, the VAM-Visual Attention Theory proposed by Schneider (1995). .

VAM theory claims that, because motor action (e.g. looking or fixating) can be directed only to one item, in order to program a saccade to a target, attention is required to be redirected to that location in space, but not all the tokens to be formed can be remembered because of the limited capacity of the short-term memory.

In short, VAM propose that attention shifts will determine what information is encoded into transsaccadic memory and hence remembered across eye movements. Attention is shifted around the visual field before the eye movements, this indicating that information from regions surrounding that of each fixation is encoded and remembered through a peripheral perception. Kosslyn (1994,1995) has also proposed an attention shift subsystem which is guided by stimulus impulses that may be detected by the peripheral vision, even “in the corner of one’s eye”(Kosslyn,1994).

Visual perception and imagery common structures and mechanisms

Perception and imagery share many common structural features, and this have been demonstrated by the interference phenomenon (Perky,1910), electrophysiological measurements (Farah,1988) and from positron emission tomography studies (Kosslyn,1994).

Finke (1985) have reviewed the existing theories relating perception to visual imagery, and he have identified three categories of such theories: structural theories that propose identical spatial and pictorial properties for perceived and imaged objects, represented by Kosslyn (1980), functional theories which propose formation and transformation of mental images contribute to object recognition and comparison, represented by Shephard & Cooper (1982), and interactive theories which propose that imagery contributes directly to ongoing perceptual processes, represented by Finke (1971) and Peterson & Graham (1974). This experiment have followed a structural approach and the theory of perception and imagery proposed by Kosslyn (1995), have been used as guide line.

Kosslyn’s computational approach model to perception and imagery

Kosslyn have proposed a retinotopically visual buffer, that is an integrated “image” is projected in the visual buffer. The visual buffer contains an attention window, that allows the attention to select the needed information present in the visual buffer, and because the visual buffer contains much more information that can be processed at a time, the attention window selects , only a part of the visual buffer at a time. For shifting the attention it is proposed a special subsystem that function both ways, both from top-down commands and from bottom-up or stimulus-based impulses. Thus the attention window is adjusted on one side by the stimulus-based attention-shifting subsystem and on the other side by the higher cortical commands zones. The output from the attention window is sent to a recognition system which allows shapes to be matched. The encoding of the visual information is understood to contain a visual component as well as a motor component, in order to identify the image and its location in space. It is conceived in Kosslyn’s model that the visual buffer and the attention window play similar roles in both perception and imagery, and the properties of each, affect both sorts of processes. It is proposed that imagery is a part of the perception itself, and that the shared structures are a support for this. In the experiment there have been considered only the visual buffer, the attention window and the spatiotopic map construction from Kosslyn’s theory.

Working definitions and hypothesis

Three groups have been selected for this experiment, and each group was defined using the window attention equivalent from Kosslyn’s theory. The first group, named Free Vision, was defined by a ”mobile” attention window but limited to a certain resolution degree area, the area supposed to be selected for further processes. The second group, named Central Fixation , was defined by a “ fixed and contracted” attention window, where the contraction was to its utmost limit, equivalent with a point seen at a distance of 60 cm. The third group, named Unfocused, was defined by a “ fixed and expanded “ attention window where the expansion was to its utmost limit, equivalent with the entire visual field. It was conceptualized that in order to demonstrate the existence of a particular visual perception style, there should be clear differences in the encoding processes , clear differences in the retrieval process, and clear differences in accuracy and vividness.

The first goal was to test the flexibility of the attention window and to identify a Holistic Visual perception style with its implications for achieving a meditative state. Theoretically, the Holistic Vision could be characterized as having the attention window expanded ad infinitum which may force the visual system into a parallel cognitive processing. The second goal, was to replicate an early experiment of Brandt & Stark (1997), used as a accepted standard for the normal visual perception. The Free Vision group was considered as a control group for the other two groups, the Central Fixation and the Unfocused, or the Holistic Visual perception .

The hypothesis for this study was:

If during the visual perception task, a particular eye movement pattern shall be observed, than during the imagery task, the same eye movement pattern shall be observed.

The predictions for this experiment were: 1) the Central Fixation group shall keep the eyes fixed in the same place during the imagery task as in the visual perception task, and their encoding shall be limited to the fixation point. The vividness shall also be expected to be higher than proposed by earlier studies (Neiser,1967, Sheehan & Neiser,1969). 2) the Unfocused group shall keep the eyes fixed in the the same region during the imagery task as in the visual perception task, and their encoding and vividness in imagery shall be the best from all groups. 3) the Free Vision group shall follow the same scanpaths during imagery task, as during the visual perception task, with a high accuracy in retrieval and a moderate vividness in the imagery task.

In order to test the hypothesis, an experiment was designed, composed by two conditions, one visual perception task and an imagery task in which the independent variable was perceptual style. It was expected that a manipulation of the visual perception style would influence the imagery style, directly reflected by the eye movement patterns , accuracy of recall and vividness of the imagery. A vividness questionnaire was also used in order to measure the so-called vividness IQ of each participant (that is, an imagery ability that is thought to be stable and general trait of a particular individual), and to assign them randomly to three balanced groups. It was also expected that the visual perceptual style would influence the accuracy of retrieval from long-term memory, and for this reason, a memory test was conducted at the end of the imagery task.

METHODS

Participants

Twelve University of Tromsø. students, nine females and three males, volunteered to participate as paid participants to this experiment which was conducted in the cognitive neuropsychology laboratory at the Psychology Institute of Tromsø. All participants reported having normal or corrected to normal vision (with contact lenses) and their age was between twenty three to forty one. The participants were told that the experiment intended to test the influence of “inspirational” landscape pictures on the memory for geometric figures. All the participants were naive about the true purpose of the experiment, and during the debriefing session at the end of the experiment, this was confirmed.

Apparatus and Materials

Eye movements were recorded by means of a remote eyetracking device using infrared light. The Remote Eye Tracking Device, R.E.D. built by SMI-SensoMotoric Instruments from Teltow-Germany, which uses iView-software. The R.E.D.-II can operate at a distance of 0,5 - 1,5m, the recording eye tracking sample rate is 50/60 Hz., and the resolution is better than 0,1 degree. The eyetracking device is operating by determining the positions of two elements of the eye: the center of the pupil and the center of the corneal reflection. The sensor, a video camera views the left eye of the participant using infrared light and the video signal is processed to extract the features of interest: the center of the pupil and the center of the corneal reflection. The coordinates of all the boundary points are fed to a computer that, in turn, determines the centroids of the two elements. The vectorial difference between the two centroids is the “raw” computed eye position. There was electric light in the room during the experiment, which was not interfering with the recording capabilities of the apparatus.

“Vividness of Visual Imagery Questionnaire”, or in short, VVIQ Survey questionnaire developed by Mark (1973) was used to determine a vividness of imagery estimate of each participant. The questionnaire consists of sixteen questions, asking the participant first to image a scene and then to rate the vividness of the mental image on a five-point rating scale. Here is an example of a question from the questionnaire:” Think of some relative or friend whom you frequently see (but who is not with you at present), and consider carefully the picture that comes before your mind’s eye. 1-The exact contour of the face, head, shoulders, and body.” The rating scale was ranging from one (“No image at all, you only “know” that you are thinking of the object”), through three (“Moderately clear and vivid”), to five (“Perfectly clear and vivid as normal vision”).

Stimuli

The pictures used as stimuli were derived from a study by Brandt and Stark (1997) . The stimuli consisted of eight , 6x6 grid irregularly-checkered diagrams, each containing five black squares . The five black squares were spread in all the directions on the diagram, with no identical patterns. The stimuli 10x10 cm., we presented on the full screen of a flat screen, color monitor with a diameter of 49 cm, in the middle of a sky-blue screen . The stimuli presentation was done using software ACDSee 32v2.4. The participant was seated in the front of the monitor at 60 cm., and with the head placed in a chin-and-forehead rest apparatus to minimize head movements, to maintain the viewing distance and the 3x3 degrees (horizontally/vertically) visual angle field of the diagram.

Figure 1. Stimulus prototype

Procedures

The participants were randomly divided between three experimental groups, four in each group. The first group was the Free Vision condition, the second group was the Central Fixation condition and the third was the Unfocused condition. Before the experiment, each participant was asked to read the instructions for the experiment, the first part consisting of general instructions common to all groups while the second part were specific instructions for each group. The experiment was divided in two tasks, one perception task and one imagery task and two ratings for vividness of visual memory and accuracy of memory. The instructions for the Free Vision group for the perception task were: ”Look carefully at the figure that shall appear on the monitor and try to remember it as precisely as possible.” The instructions for the Central Fixation group in the perception task were: “Look, and keep your eyes focused in the center of the diagram, and try to remember the whole figure as precisely as possible.” The instructions for the Unfocused group for the perception task were: “Look straight ahead, unfocused and keep an even look at the whole surface and try to remember the whole figure as good as you can.” The instructions for the imagery task were common for all the groups: “ Try to imagine the figure you just saw while keeping the eyes open.” There was no instruction about how to keep the gaze in the imagery task. After reading the experiment’s instructions, each participant was seated comfortably on a chair with his or her head placed in a chin-and-forehead rest apparatus. The eye movements were recorded twice, for 20 sec in the perception task and for another 20 sec in the imagery task, for each of the eight experiment stimuli.

The experimental protocol

After the calibration of the apparatus , the experimental session was starting immediately, asking the participant to keep the head unmoved in the support.

Each trial consisted of the following sequence of events: 1) six nature landscape pictures were presented followed by a prototype of the stimuli and an empty diagram used for the imagery task. The pictures were shown to all groups, but the “unfocused” group have received five minutes training on them to learn how to look unfocused. All the groups were then familiarized with the stimulus prototype and with the empty diagram which they would see while doing the imagery task. 2) the first stimulus diagram was presented for perception for a duration of 20 sec.3) an empty screen was presented on the monitor for 40 sec, or for a duration of repeating the instructions for the imagery task, and in order to prevent the afterimages and to ensure the encoding of the stimulus in the long-term memory. 4) the empty diagram was presented for another 20 sec and the participant was instructed to image the stimulus presented before, while keeping the eyes open. 5) a diagram with numbers in all its squares was presented on the screen for the memory test. 6) the participants were asked to rate the vividness of the image in the imagery task, using a rating scale from one to five, similarly to the VVIQ Survey’s ratings, used before, which were recorded by the experimenter in a protocol. 7) the participants were asked to indicate verbally, the numbers corresponding to the black squares they remembered from the stimulus, while the experimenter was recording the answers in a protocol. 8) the next stimulus was presented for the perception task. The whole experimental session consisted of eight stimuli.9) in the end of each session, a new calibration session was conducted, in order to ensure the reliability of the collected data. 10) a debriefing session was conducted at the end of the experiment, to ensure that the participant was “naive”, that he was not aware of the true purpose of the experiment, and to make sure that the experiment have not harmed the participant.

Figure 2. Diagram of experimental procedure.

Perception Field

Report Field

Imagery Field

Masking Field

Perception Field

Calibration and eye movement analysis

Before and after each experimental session a calibration routine was recorded using a nine calibration points(a plus sign) dividing the monitor in three equal rows and colons. The calibration was over a matrix of 3x3 points, each 10 degrees apart while the viewing distance was 60 cm. The participant was instructed to fixate each location which was sampled at a rate of 1000Hz for 100msec near the middle of this interval. These recordings served to calibrate the output of the eyetracker device against spatial position, and to analyze the eye movement trajectories for fixation points and sequence of fixation points. The calibrations of all the participants shown no changes in the head position during the experimental trial, and the quality of the recorded data was thus not affected. The calibration procedure was repeated at the end of each experiment while recording the eye-movements in order to get a visual record of the precision of the calibration procedure.

Design

A 3 (perception style: free vision / central fixation / unfocused) X 2 (Visual Task: perception / imagery ) design was used.

For the latter factor, the 6 X 6 grid was divided in 9 equal parts corresponding to separate square regions of the diagram, which for experimental reasons it was named region one through nine.

Figure 3 Regions of calculus

1	2	3
4	5	6
7	8	9

Data analysis

In order to calculate the fixation times on the different locations on the stimulus, we have subdivided the stimulus in 3 X 3 matrix of areas of interest, and called them Region 1 to 9. Simple regressions and repeated-measures analysis of variance (ANOVA) were performed for each dependent measure. A probability level of .05 was used for all statistical tests; tests with a probability level greater than .10 are not reported.

RESULTS

Relationship between perception and imagery

For each participant the eye movement rate (EMR) was calculated for the viewing and imaging tasks. The EMR is defined as the percentage of time, for each item’s recording in the perception or imagery conditions, spent on each defined region (N=9). Consequently, three simple regression analyses were performed between the percent of time spent in each region in the perception condition (the regressor) and in the imagery condition (the dependent variable). in each of the 9 defined regions of the stimulus checkerboard in order to assess whether the variables of perception and imagery were related. The regression analysis for the Free Vision condition showed a clear linear relationship between EMRs in the perception task and in the imagery task (THE DATA IS ERASSED !!!)The regression plot is shown in Figure 4.

Figure 4. Free Vision: Linear regression of EMRs for perception and imagery tasks

The regression analysis for the Central Fixation also showed a linear relationship between EMRs in the perception task and in the imagery task (THE DATA IS ERRASED!!!)The regression plot is shown in Figure 5.

Figure 5. Fixation: Linear regression of EMRs for perception and imagery tasks

The regression analysis for the Unfocussed condition showed again a clear linear relationship between EMRs in the perception task and in the imagery task (THE DATA IS ERASSED!!!) The regression plot is shown in Figure 6.

Figure 6. Unfocussed: Linear regression of EMRs for perception and imagery tasks

Table 1shows the percent of time in each region for the perception and imagery condition for each group and the respective R values .

Table 1.Means, standard deviations and correlation for EMR times for the three groups in perception and imagery tasks

( THE TABLE IS ERASSED !!!!)

In addition, a repeated-measures 3 by 2 ANOVA was performed with Perception Style ( free vision, central fixation, unfocused) as the between-subjects factor, Regions (1-9) as the within-subject factor and EMR in the imagery condition as the dependent variable. The goal of this analysis was to reveal whether each Perceptual Style would affect differently the EMR during the imagery task. Importantly, a reliable interaction effect was found between Perceptual Style and Regions,(THE DATA IS ERRASED!!!). This interactive effect is shown in Figure 7. Figure 7. Interaction plot between Perceptual Styles and the EMR in the 9 Regions

In the table 2 there is a presentation of the numeric values of the means for EMR times across the 9 Regions for all the three groups during the perception and imagery tasks. The numbers indicate clearly, the near- identical EMRs times during the two tasks.

Table 2.Means for EMR times across the 9 Regions for the three groups in perception and imagery tasks

(THE TABLE IS ERRASED!!!)

In the Figure 8, the EMR recordings from the three conditions during both perception and imagery are shown to better illustate the close correlation between the two.

Accuracy of recall

A repeated-measures ANOVA was performed with Perceptual Style (Free Vision, Central Fixation, Unfocussed) with accuracy of recall as the dependent variable. This analysis showed that the Free Vision (THE DATA IS ERASSED!!!) and Central Fixation (THE DATA IS ERASSED!!!) groups were equally accurate, whereas teh Unfocussed group was significantly less accurate than these two (THE DATA IS ERASSED!!!), as also confirmed by post-hoc Scheffe’s tests (p < 0.01 in both cases). The means and SDs of the accuracy rates in each condition are shown in the table 3.

Table 3 Means and standard deviations for total correct answers for the three groups

(THE TABLE IS ERRASED!!!)

A percentage rating have also been calculated, showing that the most accurate group was the Central Fixation with (THE DATA IS ERASSED!!!)total correct answers, followed by the Free Vision group with (THE DATA IS ERASSED!!!)total correct answers and on the last place the Unfocused group with (The data is erased!!!)total correct answers.

Vividness

The scores for vividness of the imagery for each trial were analyzed in an ANOVA with Perceptual Style as the between-subjects factor. This analysis revealed no significant difference between the three groups in this «global» measure of vividness (F< 1).

A simple regression analysis between the vividness score at the VVIQ Survey questionnaire and the mean vividness ratings during the imagery task, found only a low positive regression(the data is erased!!!) but the correlation was extremely low(THE DATA IS ERASSED!!!). Means and standard deviations for vividness ratings in the imagery task is shown in the table 4.

Table 4 Means and standard deviations for vividness ratings during the imagery task

(THE TABLE IS ERRASED!!!)

Accuracy related to vividness

Both measures of vividness were expected to predict accuracy for the visual memory, therefore two simple regression analyses (collapsed across Perceptual Style groups) between these two measures was performed. Surprisingly, a regression analysis between VVIQ and accuracy showed no clear positive relation was found between these two variables (THE DATA IS ERASSED!!!). Similarly, a positive relation but too small was shown by the regression analysis of vividness in each imagery trial and accuracy (THE DATA IS ERRASED !!!!).

However, when separate regression analyses were performed for each Perceptual Style group it became clear that the Free Vision group indeed showed a clearly positive relation between vividness in imagery and accuracy variables (THE DATA IS ERASSED!!!!).

DISCUSSION

The purpose of this study was to investigate how the differences in the visual perception styles influences the visual imagery styles. Towards this end, EMRs were recorded, an accuracy for recall test and vividness in imagery ratings were administered. The hypothesis was confirmed: the same eye movement patterns during visual perception in each group were observed during the visual imagery task. In addition, the experiment did replicate the findings of an earlier study by Brandt & Stark (1997), by showing that the scanpaths during imagery are the same as in direct visual perception.

The first prediction, that the Central Fixation group shall keep the eyes fixed in the same place during the imagery task as in the visual perception task, and their encoding shall be limited to the fixation point was confirmed partially. What was confirmed, was that the eye movement pattern was identical during the perception and imagery being in fact, fixed on the same spot. This involuntary fixation of the gaze during imagery, has not been reported in other studies. An explanation for this phenomena has been proposed by Kosslyn (personal communication) for this experiment, suggesting that efferent commands to shift attention may be stored along the representation, during encoding, they can be also retrieved during recall. Gauthier and colleagues (1990) have proposed the active role of the ocular muscle proprioception in both afferent and efferent signals in controlling the vision. Bridgeman and Stark (1991) have replicated the study done by Gauthier and colleagues (1990), but could find only a week extraretinal proprioceptive information provided by the muscle spindles. There have been a long dispute between the inflow theory, or proprioception proposed by Sherrington (Gregory, 1998) and the outflow theory proposed by Helmholtz (Gregory,1998), each proposing a different direction of the fed back signals from the eye muscles for visual location of targets. The current position of Gauthier and colleagues (1990) and of Steinbach (1987), is that there is a double fed back signals from the eye muscles; the major component of the eye-position-coding signal is derived from both eye muscle activation and ocular muscle proprioception that may account up to 32% (Gauthier et.al., 1990). Therefore, it might be safe to advance an inference based on the findings of this study, that indeed, the kinesthetic information is parallel encoded with the visual information, and at retrieval this may be reenacted as a specific eye movement pattern. The prediction for this group concerning the limited encoding caused by the narrowing of the gaze area supposedly followed by the attention window was not confirmed. This fact, clearly points towards the bad coordination, between the fixation of the gaze and the adjustment of the attention window on the spot of the gaze. It has been proposed that the attention window follows the eye gaze ( Schneider, 1995), but it can also detect external stimuli from the peripheral vision ( Kosslyn, 1994). According to this logic, if the attention window is there to select the amount of information for processing , than it should control the flow of information also in a restrictive way, by inhibiting other areas in the visual buffer that are not attended to. This is clear not the case here, the participants remember all too much, and even if their attention window was supposed to be restricted to near zero. But, the fault, so far, is not with the cognitive mechanism, but with the participants’ attentional mastery skills that needs to be trained. Mahabharata epic from the indian cultural tradition, offer an illustration of this point when it is told that the master archer Arjuna could see only the eye of the bird to be pointed at, and not other parts of the bird’s body. Herigell (1974) reports of similar results in his training in the art of japanese archery. It can be that the participants could not maintain their attention window to the limited area of the fixation point because of lack of visual perceptual skill. It seems that even if you have the cognitive structure, this doesn’t become operative, unless you develop it consciously. Even if the evidence from this study in this condition is pointing against the presence of the attention window, it is believed that the fault may stay with the participants who were unable to follow the experimental instructions.

The vividness predicted for the Central Fixation group was to be good, against the findings of previous studies. Vividness for the Central Fixation group was second in the ratings between the three groups, supporting thus the prediction, even if several studies have proposed a high rate for eye movements in order to obtain a vivid image (Neisser, 1967; Sheehan & Neisser, 1969).

The second prediction, that the Unfocused group shall keep an immobility of the gaze, steady maintained on a particular spot during both the visual perception and the imagery tasks have also been confirmed. This involuntary, steady fixation on the same spot during both visual perception and imagery task, has not been investigated in other studies. An explanation for this kinesthetic phenomena, might be linked to the previous research from the literature ( Gaauthier et al.,1990; Steinbach, 1987). In the Central Fixation condition, the attention window was expected to contract to a resolution degree near to zero, but in the Unfocused condition, the attention window was expected to expand its resolution degree to the entire visual field. If we can equate the direction of the gaze with the attention window during imagery, than we might infer that in the case of the Unfocused or holistic visual perception, the cognitive processing must be parallel. In this condition, there is no evidence for any scanpaths , which might imply a serial processing, while the participants remember well the encoded information from a large resolution degree. Their visual attention was equally spread over the entire visual field, with the experimental stimulus covering only a tiny part of it, so, their accuracy of recall must be seen in the context of the whole range of information in the visual field, and by this standard, their accuracy was very high. Several other studies have indicated that a recognition of an image was prior to the scanning the image (Norton & Stark, 1971), which might indicate that cognitive parallel process are involved. From the studies of eidetic memory, there is evidence that there are no eye movements, and that the image is processes in its entirety (Hebb,1968; Haber & Haber; 1964, Luria, 1968). The attention window, which is supposed to be limited in resolution, as proposed by Kosslyn (1980,1994,1995) doesn’t support the findings for the Unfocused group which could perceive simultaneously the entire visual field. The experimental instruction for the Unfocused group was just to keep an equal spread attention over the entire visual field and to not fixate on anything in particular. During the memory test for the irregularly-checkered diagram, the five black squares were chosen out of a 200 degree of resolution , implying that the participants must have encoded the entire field in order to search after the supposed targets. In consequence, it is proposed an adjustable quality for the attention window, contrary with the current theories of visual attention ( Kosslyn, 1994; Schneider, 1995).

In order to use efficient the attention window expanded to the entire visual field, it is proposed the necessity for training of this cognitive skill. There is anecdotal evidence (Bankei,1984; Deshimaru, 1983), about the availability of this skills in perception, perception are not only restricted to visual perception, but to all perception senses. The test for visual memory for this group, found, contrary to expectations, that the retrieval was poor, which might be interpreted either as a less effective encoding style or a lack of appropriate cognitive skills. It may also be interpreted in support for the view that a normal size of the attention window is important for the efficient encoding in memory.

One prediction that was supported by the finding, was the vividness in the Unfocused group was the highest of all groups. This might be explained by the supposed used of parallel processes during both encoding and retrieving. The spread activation of parallel processes might exceed in power simple serial activation, and this may explain the increased in vividness of the imagery. It is proposed that the vividness is connected with the power of the brain activated during an imagery task, than with the number of eye movements which was supposed to contribute to the increase of vividness (Neisser,1968). Absence of eye movements have been linked to a vividness for visual images (Marks,1973), so it may be the quality of the cognitive processes, namely parallel processing, than the simple eye movements per see, which may contribute to the generation of a vivid image.

The proposed holistic visual perception, might be identified as a visual perceptual style that uses the expansion of attention window to the entire visual field, which can be monitored, and processed with accuracy. The proposed holistic visual perception is supposed to include parallel processing, which might increase the power of the mind during imagery, with consequent increasing in vividness quality. There is increased need for developing of new cognitive skills, and the proposed visual holistic perception might be one of them . Further research must consider the possibility of the training of these skills, to learn to concentrate the attention window on the spot of the gaze, skill that may require years of hard practice, but as annedoctical evidence suggests, these skills are worth the effort.

The third prediction, that the Free Vision group shall make the same scanpaths during the visual perception and imagery tasks was confirmed. This was a replication of an earlier experiment done by Brandt and Stark (1997). In addition to the replicated study, there have been introduced two extra variables, accuracy and vividness, with the prediction that a high accuracy in retrieval and a moderate vividness were expected. The findings confirmed the predictions of accuracy and vividness. The literature (Neisser, 1967; Sheehan, & Neisser, 1969) propose a high vividness for Free Vision group, but the present experiment haven’t found this. The vividness have been proposed to be dependent of the “power of the mind” involved in the encoding and recall, and not a condition of the eye movement patterns. And again, the eye movement patterns have been proposed by this experiment to represent retrieved kinesthetic information, that have been encoded together with the visual information, this being the case in simple retrieval of encoded visual information.

The role of eye movements in visual meditation

The most common type of visual meditation contains two stages: the visual perception stage and second, the imagery or visualization stage. The present experiment have demonstrated the influence of the visual perception style on the imagery, proposing that there is a strict sequence of visual perceptual skills to be followed. For the first stage of visual meditation with mandalas and yantras, David Fontana (1991,pp.86,88) suggests:

“Try not to move the eyes, but to take in the whole pattern from the one point of vision.”

“ If you are working with a mandala or a yantra, try after a time to commit it to visual memory so completely that you see it in your mind’s eye as clearly as if it hangs on the wall in front of you.”

“Gaze at it steadily, without making any attempt to commit it to memory. Then suddenly snap the eyes shut, and hold the image behind your closed eyelids, almost as if you have taken a photograph of it.”

The second stage in visual meditation is the imagery or visualization stage that is directly influenced by the perception style during the visual perception. After development of visual skills, the meditation can start directly with the stage two as described by Fontana (1995,pp.88):

“Before very long, you will find that you are able to visualize the shape at will as soon as you close your eyes, even without first gazing at it.”

The aim of visual meditation is the stabilization of the image effect, which makes the image disappear completely living only a “blank- out” or an awareness of sheer emptiness

In order to achieve a stabilization of an image, the stimulus must fall constantly in the same place on the retina. To achieve a perfect stabilization of image in direct visual perception, complex devices are needed ( Yarbus, 1967), but to achieve it in imagery it is much more easy. This is why, all the mystical traditions, use the imagery techniques in order to achieve the stabilization of image effect in the internal space (Kosslyn, 1994).

The function of the eye movements in imagery is not clear, but the present study points toward a close connection of the attention window with the oculomotor muscles. In order to achieve a stabilization of the image in imagery, it is important that the image is kept constant in the internal space and this is done by the attention window directly connected with the oculomotor muscles.

Any saccadic movements that may appear in imagery may disrupt the stabilization of the image and forsake the goal of meditation, namely, achieving a state free of visual perceptual stimuli.

In order to don’t have eye movements during imagery, it is necessary to encode the visualized object using either Fixation or Unfocussed visual perception styles. The proposed connection between the attention window and the oculomotor muscles, makes impossible to retrieve an image without being accompanied by eye movements, if it was encoded using Free Vision with saccades. The optical nystagmus in visual perception affects the stabilization of the image, but in imagery, the effect is low.

Stabilization of image effects on the brain waves activity have been investigated by Lehmann and colleagues (1967) who observed an increase of alpha rhythm, when the image disappeared. The same effect have been reported also from the studies on “ganzfelt” defined as a homogenous visual imput(Cohen & Cadwallader, 1958; Broughton, 1992).

EEG alpha activity have been found to be influenced by emotions, mental effort, orienting reaction, attention level and actual oculomotor activity ( Mulholland & Evans, 1966; Zikmund, 1972 ). The reason for alpha blocking could be attributed to mechanisms controlling alertness ( Oswald, 1957).

The EEG studies on alpha rhythm during imagery, have been rather confusing, some finding a suppression or blocking of the EEG alpha rhythm during visual imagery (Mundy- Castle, 1957 ), others finding no blocking of the alpha rhythm ( Barratt, 1956; Zikmund, 1972). For example, Zikmund (1972) found that effortless visual imagery is accompanied by EEG alpha rhythm.

The easiest meditation experiment on the habituation effect in the visual system is to keep the eyes open and to stare straight ahead in a completely obscure room. After some minutes, the person shall lose the control over the position of the eyes, not knowing if the eyes are open or shut and a meditative state shall be reach.

The importance of this experiment to the meditation research is that it has identified that the eye movement patterns are encoded together with the image and at retrieval during imagery of the image, the same eye movement patterns are present. We have inferred that the phenomenon of stabilizing of image in the visual perception can be extended to imagery, suggesting an easier process, only in the situations when the encoding of the visualized image have been without saccades. The visual perceptual styles, such as Fixation and Unfocused are proposed to be the only possible encoding strategies in visual meditation that includes imagery. The normal visual perception, or Free Vision perception style, encodes together with the image also the saccades, which at retrieval are reenacted. The eye movements disrupts the stabilization of image during imagery, and this makes the attaining of meditative absorption- Samadhy, impossible.

The ultimate aim of meditation is a direct experience of the Reality, the transcendence of perceptual phenomena (Naranjo & Ornstein, 1974).

When during visualization, the contemplated image disappears because of the stabilization effect, what remains is the pure experience of awareness of what IS. This experience must not be interpreted, reductionistic, in the tradition of cognitive science, which equates the brain with mind, but in a transpersonal perspective which acknowledge a mind that use a brain for its purposes ( Eccles, 1994).

References

Bankei,Y. (1984).Bankei Zen. Trans. by Haskel,P. New York: Grove Weidenfeld.

Barratt, P.E. (1956). Use of the EEG in the study of imagery. British Journal of Psychology,47, pp.101-114.

Brandt, S.A., & Stark ,L.W.(1997). Spontaneous eye movements during visual imagery reflect the content of the visual scene. Journal of Cognitive Neuroscience 9:1, pp.27-38.

Bridgeman, B., & Stark, L. (1991). Ocular proprioception and efference copy in registering visual direction. Vision Research, 31, 1903-1913.

Broughton, R. (1992). Parapsychology: the controversial science. London: Rider.

Bru-sgom rGyal-ba g.yung-drung (1996).The stages of A-Khrid meditation . Trans. by Kværne,P. & Rikey, T.K. Dharamsala: Library of Tibetan Works and Archives.

Buddhaghosa, B. (1975). Visuddhimagga-The path of purification. Trans. by Bikkhu Nanamoli. Kandi: Buddhist Publication Society.

Cohen, K. M.(1981). The development of strategies of visual research. In Fisher, D. F., Monty, R. A., & Senders, J. W. (Eds.), Eye movements: cognition and visual perception (pp. 271-289) . Hillsdale, NJ: Erlbaum.

Cohn, W. & Cadwaller, T. C. (1958). Cessation of visual experience under prolonged uniform visual stimulation. American Psychologist,13, 410 (abstract).

Dancu,I. & Dancu,D. (1982). Romanian icons on glass. Detroit: Wayne State Univ.

Demarais, A., & Cohen, B. (1998). Evidence for image-scanning eye movements during transitive inference. Biological Psychology 49, 229-247.

Deshimaru, T. (1983).The Zen way to the martial arts. New York: E.P.Dutton.

Eccles, Sir J.C. (1994). How the Self controls the brain. Berlin: Springer Verlag.

Farah, M. (1985). Psychophysical evidence for a shared representational medium for mental images and percepts. J.exp.Psychol.: Gen. 114, 91-103, 1985.

Farah, M. (1988). Electrophysiological evidence for a shared representational medium for visual images and visual percepts. Journal of Exp. Psyh.: General Vol.117, No.3, 248-257

Farah, M. (1995). Current issues in the neuropsychology of image generation .Neuropsychologia Vol 33, No.11, pp.1455-1471, 1995.

Feurstein, G. (1998). Tantra- the path of ecstasy. Boston: Shambhala.

Finke, R. (1985). Theories relating mental imagery to perception. Psychological bulletin, 98, 236-259.

Fontana, D. (1995). The elements of meditation. Shaftesbury: Element Books.

Gauthier, G. M., Nommay, D., & Vercher, J-L. (1990). Ocular muscle proprioception and visual localization of targets in man. Brain113, 1857-1871.

Gibson, J.J. (1979). The ecological approach to visual perception. Boston: Houghton Mifflin.

Goldenberg, G., Podreka, I., Steiner, M. & Willmes, K. (1987). Patterns of regional cerebral blood flow related to memorizing of high and low imagery words-An emission computer tomography study. Neuropsychologia, 25, 473-485.

Gregory, R. L. (1998). Eye and brain. Oxford University Press.

Haber, R. N., & Haber, R. E. (1964). Eidetic imagery: I. Frequency.Perceptual and Motor Skills, 19, 131-158.

Hebb, D. O. (1968). Concerning imagery. Psychological Review Vol75, No.6, 466-477.

Herigell, E. (1974). Zen in the art of archery. New York: Random House.

Henderson, J. (1994). Two representational systems in dynamic visual identification. Journal of Experimental Psychology: General Vol 123, No.4, 410-426.

Hirai, T. (1989). Zen meditation and psychotherapy. Tokyo: Japan Publications Inc.

Hochberg, J. (1964). Perception. Englewood Cliffs, NJ: Prentice-Hall, Inc.

Irwin, D. E. (1991). Information integration across saccadic eye movements.Cognitive Psychology, 23, 826-839.

Irwin, D. E. (1992). Memory for position and identity across eye movements. Journal of Experimental Psychology: Learning, Memory, and Cognition Vol. 18, No.2, 307-317.

Irwin, D. E. (1996). Integrating information across saccadic eye movements. Current Directions in Psychological Science, 5 , 94-100.

Irwin, D. E., & Gordon, R. D. (1998). Eye movements, attention and trans-saccadic memory. Visual Cognition,5 (1/2), 127-155.

Johari, H. (1986). Tools for Tantra. Rochester: Destiny Books.

Kahneman, D., & Treisman, A. (1984). Changing view of attention and automaticity.In R. Parasuraman & D. Davies (Eds),Varieties of attention (pp.29-61).New York: Academic Press.

Khanna, M. (1979). Yantra: the tantric symbol of cosmic unity. London: Thames and Hudson.

Kosslyn, S. M. (1980). Image and mind. Cambridge, Mass.: Harvard University Press.

Kosslyn, S. M. (1994). Image and brain. Cambridge, Mass.: The MIT Press.

Kværne, P. (1995).The bon religion of Tibet: the iconography of a living tradition. London: Serindia Publications.

Laeng, B. (1990). Le Icone. Roma: Bulzoni Editore.

Lehman, D., Beeler, G. W. & Fender, D. H. (1967). EEG responses during the observation of stabilized and normal retinal images. Electroencephalography and Clinical Neurophysiology ,22. Pp.136-142.

Luria, A. R. (1968). The mind of a mnemonist. New York: Basic Books.

Marks,D. F. (1973). Visual imagery differences in the recall of pictures. British Journal of Psychology 64, pp. 17-24.

Maurer, D., & Lewis, T. L. (1981). The influence of peripheral stimuli on infant’s eye movements. In Fisher, D. F., Monty, R. A., & Senders, J. W. (Eds.).Eye movements: cognition and visual perception. (pp. 21-31)., Hillsdale, NJ: Erlbaum.

Mulholland, T. & Evans, C. P. (1966). Oculomotor function and the alpha activation. Nature,211, pp.1278-1279.

Mundy-Castle, A.C. (1957). The electroencephalogram and mental activity. Electroencephalography and Clinical Neurophysiology,9 pp.643-655.

Naranjo, C. & Ornstein, R. E. (1974). On the psychology of meditation. New York: the Viking Press.

Noton, D., & Stark, L. W. (1971). Scanpaths in saccadic eye movements while viewing and recognizing patterns. Vision Research,11, 929-942.

Neisser, U. (1967). Cognition psychology. New York: Appleton-Century-Grofts.

Oswald, I. (1957). The EEG, visual imagery and attention. Quartely Journal of Experimental Psychology,9. pp.113-118.

Paivio, A. (1986). Mental representations: a dual coding approach. New York: Oxford University Press.

Paivio, A. (1991). Images in mind: the evolution of a theory. Exester: BPCC Wheatons Ltd.

Patanjali (1990). The Yoga sutra of Patanjali with the commentary of Vyasa. Trans. Bengali Baba. Delhi: Motilal Banarsidass Publishers.

Perky,. C.W. (1910). An experimental study of imagination. In Kosslyn, S. M. ( Eds.). Image and brain. , pp.54, 102. Cambridge, Mass: The MIT Press.

Peterson, M. J. & Graham, S. E. (1974). Visual detection and visual imagery. Journal of Experimental Psychology, 103, pp.509-514.

Rao,S.K.R. (1988). The Yantras. Delhi: Sri Satguru.

Rayner, K., McConkie, G. W., & Ehrlich S. (1978). Eye movements and integrating information across fixations. Journal of Experimental Psychology: Human Perception and Performance, Vol.4, No. 4., pp.529- 545.

Schneider, W. X. (1995). VAM: A neuro-cognitive model for visual attention, control of segmentation, object recognition, and space-based motor action. Visual Cognition, 2 (2/3), 331-375.

Shapiro,D. H., & Walsh, R. (1984). Meditation - classic and contemporary perspectives. New York: Aldine Publishing Co.

Sheehan, P. W. (1967). Visual imagery and the organizational properties of perceived stimuli. British Journal of Psychology, 58, 3 and 4, pp. 247-252.

Sheehan, P. W., & Neisser, U. (1969). Some variables affecting the vividness of imagery recall. British Journal of Psychology, 60, 1, pp.71-80.

Shepard, R. N., & Cooper, L. A. (1982). Mental images and their transformations. Cambridge Mass.: The MIT Press.

Stark, L., & Ellis, S. R. (1981). Scanpaths revisited: cognitive models direct active looking. In Fisher, D. F. , Monty, R. A. , & Senders J. W. (Eds.). Eye movements: cognition and visual perception. Hillsdale, NJ.:Erlbaum.

Steinbach, M. J. (1987). Proprioceptive knowledge for eye position. Vision Research 26, pp. 1737- 1744.

Tart, C. T. (!990). Altered states of consciousness. San Francisco: HarperCollins Publishers.

Tippett, L. J. (1992). The generation of visual images: a review of neuropsychological research and theory. Psychological Bulletin, Vol.112, No.3, 415-432.

West, M. A.( 1991). The psychology of meditation. Oxford: Clarendon Press.

Wilber, K. (1995). Sex, ecology, spirituality: the spirit of evolution. Boston: Shambhala.

Wilber, K. (1999). One taste. Boston: Shambhala.

Wilber, K. (in press). Integral psychology. Boston: Shambhala.

Yarbus, A. L. (1967). Eye movements and vision. New York: Plenum Press.

Zikmund, V. (1972). Physiological correlates of visual imagery. In Sheehan, P. W. (Eds.).The function and nature of imagery. New York: Academic Press.