- Email: [email protected]
Visual evoked potentials associated with the verbal and non-verbal problem-solving processes
BiologicalPsychology 10 (1980) 103-114 0 North-Holland Publishing Company
VISUAL EVOKED POTENTIALS ASSOCIATED WITH THE VERBAL AND NON-VERBAL
Shoji KITAJIMA,
PROBLEM-SOLVING
PROCESSES
Harumitsu MUROHASHI and Minami KANOH
Hokkaido University, Sapporo 060. Japan Accepted
for publication
1 April 1980
VEPs to six repetitive checkerboard and letter or number stimuli (presented in two triplets) were measured. The task solution process was separated into the store, retention, solve and post-solve stages. Three stimuli in intratrial stimulus positions 1, 2 and 3 were identical and another three in positions 4, 5 and 6 were also identical. The subject’s task was either naming the letter-number or matching the checkerboard stimuli in two triplets. For both stimuli, the occipital P270 was almost the same in amplitude at positions 1 and 4. During the memory retention (at position 2) the P270 enhanced for the letter-number stimulus,‘but not for the checkerboard. During the post-solve stage (at positions 5 and 6) the P270 enhanced for both stimuli. The occipital P270 enhancement seems to correlate with the decrease in the degree of significance of the task relevant stimuli.
1. Introduction
This study is concerned with the late components of the averaged visual evoked potentials (VEPs) of the human brain associated with the stimulus discrimination situation. In the typical study of the VEP correlates of the stimulus discrimination or comparison, two task relevant stimuli are presented separately in sequence. A subject must retain the memory of the first stimulus until the second one occurs which carries crucial information for the task solution. Chapman and Bragdon (1964), Shelburne (1972, 1973), Thatcher and April (1976) and Thatcher (1977) have studied the VEP correlates of the semantic information processing and demonstrated the amplitude enhancement of the VEP for the second relevant stimulus. Chapman (1973), using a sequence of four letter and number stimuli, has shown that the VEP amplitude (VEP area) was significantly more positve when the stimuli were task relevant than when the same stimuli were task irrelevant. Chapman has measured the VEP amplitudes for ‘perceive-store’ and ‘perceive-compare-solve’ programs, and found that the amplitudes of the vertex VEPs for the ‘perceive-store’ program were significantly different from those for the ‘perceive-compare-solve’ program. The experiment reported in this paper was designed to analyze the VEP behav103
104
S. Kitajima et al. / VEP and problem-sohirq
iors during the task solution process which was separated into four stages: store, retention, solve and post-solve. In this experiment, three identical stimuli were presented repetitively as the first task relevant triplet (intratrial stimulus positions from 1 to 3) in order to separate the memory process into the store (position 1) and retention (positions 2 and 3) stages. This triplet was followed by the second triplet which was composed of another three identical stimuli (intratrial stimulus positions from 4 to 6). The second triplet separated the task completion into the solve (position 4) and post-solve (positions 5 and 6) stages. Tasks were naming the letters or numbers (letter-number hereafter) and matching the checkerboard patterns between two triplets. Under this condition the retained memory was verbal for the letter-number stimulus, while it was non-verbal for the checkerboard stimulus. When identical stimuli are presented repetitively at regular intervals, an initial stimulus produces the VEP and auditory evoked potential (AEP) of the largest amplitude (Ritter, Vaughan and Costa, 1968; Butler, 1973; Kitajima, 1978). Studies on human memory have shown that an initial letter or word of a memory list is most easily remembered (Rumelhart, 1977). A subject in this study could draw a task relevant information from any stimulus in the triplets. However, these data of the initial effects suggest that the subject regards the initial stimuli in the triplets as the most significant for the task solution. In the case of the first triplet, the subject perceives and stores the stimulus in position 1. He must retain this memory until the presentation of the second triplet. During the memory retention he can consolidate the retained memory, if necessary, by means of extracting the sensory information carried by the stimuli in positions 2 and 3. These stimuli may be more useful in consolidating the non-verbal memory and less useful in consolidating the verbal one. In other words, these stimuli may be more significant for the task solution when the memory is non-verbal than when it is verbal. The second triplet measured the effects of the task completion (separated into the solve and post-solve stages) upon the VEP amplitudes. As the short-term memory of the first triplet decays as a function of distance in time from its offset, the task must be solved at the beginning of the second triplet. Therefore, the stimuli in positions 5 and 6 (the post-solve stage) become less significant (or more redundant) for the matching task as well as for the naming one. Hypotheses to be proved in this experiment are: (1) the VEP amplitudes for the letter-number and checkerboard stimuli are maximum at intratrial stimulus positions 1 and 4; (2) the VEP amplitude for the checkerboard stimulus is larger than the amplitude for the letter-number one at positions 2 and 3; and (3) the VEP amplitudes for both stimuli become smaller at positions 5 and 6.
S. Kitajima et al. / VEP and problem-solving
105
2. Methods 2.1. Subjects Subjects were 11 college students, seven males and four females. They ranged in age from 21 to 27, had normal or corrected vision and no history of neurological or visual defects. 2.2. Stimulus presentation Three patterns were used (fig. 1). They were composed of alternating light and dark squares and rectangles on the same-sized checkerboard with 6 X 6 or 12 X 12 cell units. A 6 X 6 cell units complete checkerboard served as a warning stimulus (WS). An incomplete checkerboard composed of six light squares served as a task stimulus. For each column of a 6 X 6 checkerboard, only one cell was light, and the vertical position of the light cell was chosen at random. Another task stimulus was a letter (from captal A to Z, for six subjects) or number (0 and 2 to 9, for five) composed of light cells on a 12 X 12 checkerboard. These task stimuli were matched in total luminance. The stimuli subtended a visual angle of ab’out 3”. Stimulus luminance was 0.05 ml and exposure duration was 300 ms. The stimuli were
WARN ING
ST IMIJLIJS
CHECKERBOARD
LETTER
NUMBER
Fig. 1. Stimulus patterns. A fixation point (not shown) was located at the center terns. Each stimulus was exposed as a light figure against a dark background.
of the pat-
106
S. Kitajima et al. 1 VEP and problem-solving
projected on a translucent screen by a Kodak incandescent projector with a mechanical shutter mounted outside the experiment room. The shutter was activated by an electric pulse, rise and fall time were about 20 ms. There was no shutter noise in the experiment room. Two trials were used: a checkerboard trial and a letter or number trial. The checkerboard and letter trials were used for six subjects, while the checkerboard and number trials for the other subjects. Each trial was composed of seven successive stimuli: the WS, the first triplet consisting of three identical stimuli (intratrial stimulus positions 1,2 and 3) and the second triplet consisting of another three identical stimuli (intratrial stimulus positions 4, 5 and 6). If the frist triplet was composed of either checkerboards, letters, or numbers, the second one was composed of either checkerboards, letters, or numbers, respectively. Fifty checkerboard and 50 letter or number trials were presented in four blocks. About half of 25 trials in the block were the checkerboards and another half were letters or numbers which were arranged at random. The interstimulus interval (ISI) between onsets of the WS and the stimulus in position 1 was 1500 ms. The IS1 between the stimuli in positions 3 and 4 was 1500 ms. The stimuli in each triplet were separated with the IS1 of 1000 ms. The intertrial interval was varied from 20 to 25 s with a mean of 23. The stimulus duration and ISIS were controlled by a preset digital timer. 2.3. Procedure
The subject was seated comfortably in a reclining dentist’s chair and faced a screen in a darkened, sound-proof and electrical shielded room. After becoming fully adapted to the darkness, he was instructed to look binocularly at a faint red fixation point during the stimulus presentation. The centers of the stimuli coincided with the fixation point. The subject was required to perform two tasks. If the stimulus was a checkerboard, he had to make a pattern matching between the first and second triplets and to answer as to whether the triplets were the same or different. If the stimulus was a letter or number, he had to name letters or numbers in the triplets. To avoid myogenic effects on electroencephalographic (EEG) responses, he was asked to answer about 1 s after the termination of the trial. After four practice trials, 100 trials were presented in four blocks. An interval between the blocks was about 3 min, serving as a short rest. During the rest period, a stimulus-mounted tray was changed with another. A few seconds before the presentation of the block, the subject was signalled to look at the fixation point by a brief tone of 1 kHz. After the block presentation, he was signalled to relax by another tone. 2.4. Recording system and data analysis EEG responses were recorded from frontal, temporal, vertex and occipital scalp locations (F3, F4, C3, C4, Cz, 01, and 02 in the international lo-20 system),
S. Kitajima et al. / VEP and problem-solving
107
referenced to linked ears, using chlorided silver disk electrodes. Interelectrode resistances were less than 5 kfi. Eye movements were recorded from electrodes placed on the lateral canthus and the supercilium. EEG and electrooculographic (EOG) responses were amplified by a San-ei 14-channel electroencephalograph (time constant: 0.3 s) and recorded by a TEAC 14-channel FM magnetic tape recorder. EEG responses contaminated by electromyographic (EMG) responses and large eye movements were discarded. Forty EEG responses to the checkerboard trials were averaged using a minicomputer HP-2100 at an analysis time of 3 s (giving a resolution of 7.8 ms/point). Forty EEG responses to the letter or number trials were also averaged in the same way. Averaged EOGs and VEPs were plotted with a Riken X-Y plotter. The peak amplitude was measured with respect to a baseline chosen as the mean voltage over the first 40 ms of the post-stimulus EEG responses. The statistical significance of the amplitude differences between the tasks and across the stimulus positions was evaluated by means of the analysis of variance and the Student’s t-test for paired observations.
3. Results No subjects failed to solve the tasks. In the post-experiment interview, they reported the strategies which they employed to solve the tasks. The stimuli in intratrial stimulus positions 1 and 4 were regarded by all subjects as the most important and looked at most carefully. The matching task was reported as more difficult than the naming. There were no significant differences between the VEP behaviors for the letter and number stimuli. They varied with changes in the stimulus position in the same way. Therefore, these VEP data were joined together in the following analysis. VEP waves recorded from seven electrode loci for a subject are seen in fig. 2. For the VEPs recorded from the centro-frontal five electrodes (F3, F4, C3, C4 and Cz) the most negative peak between 110 and 130 ms, with a mean latency of 120, after the stimulus onset was termed N120, the most positive one between 170 and 200 ms, with a mean latency of 180, was termed P180, and the most positive one between 300 and 460 ms, with a mean latency of 350, was termed P350. For the VEPs recorded from the occipital electrodes (01 and 02), the most negative peak between 160 and 200 ms, with a mean latency of 180, was termed N180, and the most positive one between 250 and 290 ms, with a mean latency of 270, was termed P270. The amplitudes of these peaks recorded from the centrofrontal loci are seen in fig. 3(a), and those recorded from the occipital loci in fig. 3(b). These amplitudes were averaged values across 11 subjects for Cz and 02, while across 10 for C3, C4, and 01, and across nine subjects for F3 and F4, because of EMG and EOG contaminations and poor electrode condition. The N120 amplitudes at the centro-frontal loci were not significantly different either across the stimulus positions or between
’
30-78 ,,.. 8-
1
500 m*ec
POSITION 2
500miec
POSITION
3
LETTER
-NUMBER
4
iO0 inrec MAT&J VJo~‘~50~n*ci~ 8
POSITION
CHECKERBOARD POSITION
5
‘50imw
POSITION 6
I‘ip. 2. VLP waveforms for a subject. tach VIP wils obtained through averaging 40 1 1% responses recorded monopolarly from the centro-frontal (F3. 1’4. C3. C4 and Cz) and occipital (01 and 02) electrodes with reference to the linked ears. The electrode IOCIare wzn on the left and middle ordinates. The VtPs to the letter-number stimulus (the number for this subject) wxe traced with solid lines, whereas the VEPs to the checkerboard stimulus were traced with dotted lines. Positions from 1 to 6 stand for the intratrial stimulus positions. Latencies in msec are seen on the abscissa. The stimuli were presented at 0 ms and terminated at 300 ms. In the case of this subject, the latencies of the centro-frontal N120, P180 and ~350 were about 120, 180 and 300 ms respectively, and those of the occipital NlXO and P270 were about 180 and 270 ms, respectively. The ccntro.frontal PI80 was maximum in amplitude at position 1 and P350 was maxunum in amplitude at posltions 1 and 4. The amplitudes decreased 3c~oss the stimulus positions. The occipital P270 enhanced (more positive) at position 2 in the flr$t triplet for the letter-number stimulus. but not for the clleckerboard stimulus. At position 5 in the second triplet, the P270 enhanced (more pusitwcl for the letter-number and chcckcrboard
MATS;1
POSITION
-
S. Kitajima et al. / VEP and problem-solving Chn
10A’ 6
P350
P180
*+
A I3
N120
109
.*
.
Cm-11
A
‘*
P350
8
If
P180
A
\
B
**
._N 120
01 n=lO P270
A:.10
-
LN
02
B:
_.---
CK
P270
n=ll
4:* B:
8 -
12 .
-4 -
3 .
.
45 .
.
6 -
12
*
*
3
.
45
.
.
6
.
B:*
-6 Fig. 3. The VEP amplitudes measured from the baseline for the letter-number (LN) (traces with solid Lines) and checkerboard (CK) (traces with dashed lines) stimuli. The amplitude values are plotted as a function of the intratrial stimulus positions from 1 to 6 which are seen on the abscissa. Fig. 3(a) shows the amplitudes of the N120, P180 and P350 of the VEPs recorded from the centro-frontal (F3, F4, C3, C4 and Cz) electrodes, and fig. 3(b) shows the amplitudes of the N180 and P270 of the VEPs recorded from the occipital (01 and 02) electrodes. The amplitude values were averaged across 11 subjects for the VEPs from Cz and 02, while across 10 for C3, C4, and 01, and across nine subjects for F3 and F4. The amplitudes in ~.IVare seen on the left ordinate. The capital letters A and B indicate the degree of the significance of the amplitude differences revealed by the analysis of variance. A is for the difference between the letter-number and checkerboard stimuli, and B is for the difference across the stimulus positions (*: p < 0.05; **: p < 0.01). An asterisk between the amplitude values for the letter-number and checkerboard stimuli shows that the student’s t-test revealed significant differences between these amplitudes 0, < 0.05). As seen in fig. 3(a), the amplitudes of the P180 at the centro-frontal loci were maximum at position 1 and decreased across the stimulus positions. The P350 amplitudes were largest at positions 1 and 4, and decreased at the other stimulus positions. At position 4, the P350 to the checkerboard stimulus was larger in amplitude than the P350 to the letter-number stimulus. As seen in fig. 3(b), the occipital N180 amplitudes were maximum at positions 1 and 4, and decreased across the stimulus positions. The P270 amplitudes increased at position 2 for the letter-number stimulus but not for the checkerboard stimulus. The P270 amplitudes at position 5 for the letter-number stimulus were significantly larger than those for the checkerboard stimulus. Though not significant, the P270 for the checkerboard stimulus increased in amplitude across the stimulus positions from 4 to 6. A: 0.10 in 01 indicates p < 0.10.
110
S. Kitajima et al. / VEP and problem-solviq
the letter-number and checkerboard stimuli. The P180 amplitudes at these loci were largest for the stimulus in position 1. The analysis of variance revealed that the decreases of the P180 amplitudes across the stimulus positions were significant: for five loci [F(5, 96) =3.79, p
S. Kitajima et al. / VEP and problem-solving
111
post-solve stages) upon the VEP behaviors were analyzed. The behaviors of the centro-frontal (F3, F4, C3, C4 and Cz) P350 proved the hypotheses that the VEP amplitudes are maximum at intratrial stimulus positions 1 and 4 and become smaller at positions 5 and 6, but did not prove the hypothesis that the VEP amplitude for the checkerboard stimulus is larger than the amplitude for the letter-number one at positions 2 and 3. The behaviors of the occipital P270 did not prove any of them. The enhancement of the centro-frontal P350 was observed for the store and solve stages (at positions 1 and 4). This is consistent with reports of the late positive wave (P300 or P3 in some reports) in relation to the significance of the task relevant stimulus (Smith, Donchin, Cohen and Starr, 1970; Harter and Salmon, 1972; Ritter, Simson and Vaughan, 1972; Wilkinson and Lee, 1972;Picton and Hillyard, 1974; Rohrbaugh, Donchin and Eriksen, 1974; Courchesne, Hillyard and Galambos, 1975; Squires, Donchin, Herning and McCarthy, 1977; reviews NIXmen, 1975; Sabat, 1978), and seem to correspond with the subjects’ reports that they regarded the stimuli in positions 1 and 4 as most significant for the task solution. Chapman (1973) using letter and number stimuli, has reported that the vertex VEP areas (mean amplitudes in his term) during 480 ms after the stimulus onset were significantly different between the store and solve stages. In the present study, the vertex P350 amplitude for the checkerboard stimulus was different between the store and solve stages, but the P350 for the letter-number stimulus was not different between these stages. The amplitude of the human VEP is influenced by the physical and temporal stimulus properties and subjective psychophysiological factors. Many studies on recovery processes of human VEPs have demonstrated that the recovery of the VEP amplitude is dependent upon an interstimulus interval (1%) (Ciganek, 1964; Schwartz and Shagass, 1964; Spehlmann, 1965; Vasconetto, Floris and Morocutti, 1971; Davis, Osterhammel, Wier and Gjerdingen, 1972; Lehtonen, 1973; Kitajima, Morotomi and Kanoh, 1975; Kitajima, 1978). The amplitude decreases of the centro-frontal P180 at positions 2, 3, 5 and 6, the slight increase at position 4, and the amplitude changes of the occipital N180 across the stimulus positions can be explained in terms of the recovery processes. Chapman (1973) has reported that the vertex VEP areas during 225 ms after the stimulus onset decreased as a function of the stimulus positions. An interesting finding in this study was the amplitude increases of the occipital P270 during the retention stage. This VEP enhancement was observed for the letternumber stimulus, bot not for the checkerboard stimulus. Another finding was the amplitude increase of the same P270 during the post-solve stage which.was observed for both stimuli. The physical properties of the stimulus could contribute to the occipital P270 enhancement. MacKay and Jeffreys (1973) Spekreijse, van der Tweel and Zuidema (1973), Jeffreys (1977) and Spekreijse, Estevez and Reits (1977) have demonstrated that different pattern stimuli can produce different amplitudes of the occip-
112
S. Kitujinu et al. / VEP and prohkm-solving
ital VEPs. The differences in the physical properties between the letter-number and checkerboard stimuli may explain the differences between the VEPs to these stimuli, It was observed in this experiment that the occipital P270 alnplitudes for the letter-number stimulus in position 1 (the store stage) were almost the same as those for the checkerboard stimulus in the same position. Therefore, it seems unlikely that the physical properties can explain why the P270 increased significantly in amplitude at position 2 (the retention stage) only for the letter-number stimulus. The amplitude increase of the occipital P270 during the post-solve stage (at positions 5 and 6 in the second triplet), though the P270 enhatl~ement for the checkerboard stimulus was not significant, suggests that the redundancy (a decrease of an amount of necessary information) of the stimulus for the task solution may be one of the factors which contributed to the P270 enhancement. As the short-term memory of the first triplet decays as a function of distance in time from it’s offset, the tasks must be solved at the beginning of the second triplet. The subjects’ reports and the enhancement of the centro-frontal P350 at ~~osition 4 support the hypothesis that the tasks were completed most probably at this stimulus position. After this solve stage, the stimuli in positions 5 and 6 might become redundant for the task solution. It is possible to correlate the occipital P270 enhancement during the post-solve stage with the progress of the task-redundancy at these stimulus positions. In the case of the first triplet, the vertex P350 amplitude and the subjects’ reports seem to support the hypotheses that the matching task was more difficult than the naming one and that during the retention stage the checkerboard stimuli were more significant than the letter-number ones. The enhancement of the occipital P270 seems to correlate with the less-significance or task-redundancy of the letter-number stimulus during the retention stage. There are reports which have demonstrated that the AEP and VW do not habituate to repetitive stimuli if they are significant for the task solution (Sutton, Tueting, Zubin and John, 1967; Courchesne, Hillyard and Courchesne, 1977; reviews Callaway, 1973; Picton, Hillyard and Galambos, 1976). The VEPs to repetitive stimuli should become smaller in amplitude if these stimuli are task-redundant. The VEP enllarl~~ment reported in this paper seems to be inconsistent with the habituation principle. The repetition of the identical task relevant stimuli and the decrease in the degree of significance of these stimuli may contribute to this enhancement. Further analysis of this problem is in progress in our laboratory.
References Butler, R.A. (1973). The cumulative effects of different stimulus repetition rates on the auditory evoked response in man. Electrocncephalography and Clinical Ncurophysiology, 35, 337-345. Callaway, E. (1973). Habituation of averaged evoked potentials in man. In: Pecke, H.V.S. and
S. Kitajima et al. / VEP and problem-solving
113
Herz, M.J. (Eds). Habituation Vol. II. Physiological Substrates. Academic Press: New York, 153-174. Chapman, R.M. and Bragdon, H.R. (1964). Evoked responses to numerical and non-numerical visual stimuli while problem solving. Nature (London), 203, 1155-1157. Chapman, R.M. (1973). Evoked potentials of the brain related to thinking. In: McGuigan, F.J. and Schoonover, R.A. (Eds). The Psychophysiology of Thinking. Academic Press: New York, 699108. Ciganek, L. (1964). Excitability cycle of the visual cortex in man. Annals of New York Academy of Science, 112,241-253. Courchesne, E., Hillyard, S.A. and Courchesne, R.Y. (1977). P3 waves to the discrimination of targets in homogeneous and heterogeneous stimulus sequences. Psychophysiology, 14,590597. Courchcsne, E., Hillyard, S.A. and Galambos, R. (1975). Stimulus novelty, task relevance and the visual evoked potential in man. Electroencephalography and Clinical Neurophysiology, 39,131I143. Davis, H., Osterhammel, P.A., Wier, C.C. and Gjerdingen, D.B. (1972). Slow vertex potentials: Interactions among auditory, tactile, electric and visual stimuli. Electroencephalography and Clinical Neurophysiology, 33,537-545. Harter, M.R. and Salmon, L.E. (1972). Intra-modality selective attention and evoked cortical potentials to randomly presented patterns. Electroencephalography and Clinical Neurophysiology, 32,605-613. Jeffreys, D. (1977). The physiological significance of pattern visual evoked potentials. In: Desmedt, J.E. (Ed.). Visual Evoked Potentials in Man: New Development. Clarendon Press: Oxford, 134-167. Kitajima, S. (1978). The cumulative inhibitory effect of repetitively flashed stimuli on the recovery process of the human visual evoked potential to a test stimulus. Electroencephalography and Clinical Neurophysiology, 44, 364-372. Kitajima, S., Morotomi, T. and Kanoh, M. (1975). Enhancement of evoked responses to brief flashes after offset of preexposed light stimulation: A critical moment. Vision Research, 15, 1213-1216. Lehtonen, J.B. (1973). Functional differentiation between late components of visual evoked potentials recorded at occiput and vertex: Effect of stimulus interval and contour. Electroencephalography and Clinical Neurophysiology, 35,75-82. MacKay, D.M. and Jeffreys, D.A. (1973). Visually evoked potentials and visual perception in man. In: Autrum, H., Jung, R., Loewenstein, W.R., MacKay, D.M. and Teuber, H.L. (Eds). Handbook of Sensory Physiology Vol. VII/3B. Springer-Verlag: Berlin, 647-678. N%&ren, R. (1975). Selective attention and evoked potentials in humans - A critical review. Biological Psychology, 2,237-307. Picton, T.W. and Hillyard, S.A. (1974). Human auditory evoked potentials. II. Effects of attention. Electroencephalography and Clinical Neurophysiology, 36, 191-200. Picton, T.W., Hillyard, S.A. and Galambos, R. (1976). Habituation and attention in the auditory system. In: Keidel, W.D. and Ncff, W.D. (Eds). Handbook of Sensory Physiology Vol. 5. Springer-Verlag: Berlin, 343-387. Ritter, W., Simson, R. and Vaughan, H.G. Jr. (1972). Association cortex potentials and reaction time in auditory discrimination. Electroencephalography and Clinical Neurophysiology, 33, 547-555. Ritter, W., Vaughan, H.G. Jr. and Costa, L.D. (1968). Orienting and habituation to auditory stimuli: A study of short term changes in average evoked responses. Electroencephalography and Clinical Neurophysiology, 25,550-556. Rohrbaugh, J.W., Donchin, E. and Eriksen, C.W. (1974). Decision making and P300 component of the cortical evoked response. Perception and Psychophysics, 15, 368-374.
114
S. Kitajirna et al. / VEP and problem-solving
Rumelhart, D.E. (1977). Introduction to Human Information Processing. John Wiley, New York. Sabat, S.R. (1978). Selective attention, cortical evoked responses, and brain function in human subjects: A critical review and theory. Biological Psychology, 7, 175~-201. Schwartz, M. and Shagass, C. (1964). Recovery functions of human somatoscnsory and visual evoked potentials. Annals of New York Academy of Science, 112, 510-525. Shelburne, S.A. Jr. (1972). Visual evoked responses to word and nonsense syllable stimuli. Electroencephalography and Clinical Neurophysiology, 32, 17--25. Shelburne, S.A. Jr. (1973). Visual evoked responses to language stimuli in normal children. Electroencephalography and Clinical Neurophysiology, 34, 135-143. Smith, D.B.D., Donchin, I:., Cohen, L. and Starr, A. (1970). Auditory averaged evoked potcntials in man during selcctivc binaural listening. Electroencephalography and Clinical Neurophysiology, 28, 146-152. Spehlmann, R. (1965). The averaged electrical rcsponscs to diffuse and to pattcrncd light in human. Electroencephalography and Clinical Neurophysiology, 19, 560 --569. Spckreijse, H., Estevez, 0. and Rcits, D. (1977). Visual evoked potentials and the physiological analysis of visual processes in man. In: Dcsmedt, J.11. (Ed.). Visual Evoked Potentials in Man: New Dcvclopments. Clarendon Press: Oxford, 16-89. Th. (1973). Contrast evoked responses in Spekreijse, H., van der Twecl, L.H. and Zuidcma, man. Vision Research, 13, 1577-1601. Squires, K.C., Donchin, E., Her&g, R.I. and McCarthy, G. (1977). On the influence of task relevance and stimulus probability on event-related-potential components. Elcctrocncephalography and Clinical Neurophysiology, 42, l-14. Sutton, S., Tucting, P., Zubin, .I. and John, E.R. (1967). Information delivery and the sensory evoked potential. Scicncc, 155, 1436-1439. Thatcher, R.W. (1977). Evoked-potential correlates of delayed letter-matching. Behavioral Biology, 19, l-23. Thatcher, R.W. and April, R.S. (1976). Evoked potential correlates of semantic information processing in normals and aphasics. In: Richer, R.W. (Ed.). The Neuropsychology of Language. Plenum Press: New York, 95-124. Vasconetto, C., Floris, V. and Morocutti, C. (1971). Visual evoked responses in normal and psychiatric subjects. Electrocncephalography and Clinical Neurophysiology, 3 1, 77 --83. Wilkinson, R.T. and Lee, M.V. (1972). Auditory evoked potentials and selective attention. k:lectroencephalography and Clinical Neurophysiology, 3 3,4 11-4 18.
VISUAL EVOKED POTENTIALS ASSOCIATED WITH THE VERBAL AND NON-VERBAL
Shoji KITAJIMA,
PROBLEM-SOLVING
PROCESSES
Harumitsu MUROHASHI and Minami KANOH
Hokkaido University, Sapporo 060. Japan Accepted
for publication
1 April 1980
VEPs to six repetitive checkerboard and letter or number stimuli (presented in two triplets) were measured. The task solution process was separated into the store, retention, solve and post-solve stages. Three stimuli in intratrial stimulus positions 1, 2 and 3 were identical and another three in positions 4, 5 and 6 were also identical. The subject’s task was either naming the letter-number or matching the checkerboard stimuli in two triplets. For both stimuli, the occipital P270 was almost the same in amplitude at positions 1 and 4. During the memory retention (at position 2) the P270 enhanced for the letter-number stimulus,‘but not for the checkerboard. During the post-solve stage (at positions 5 and 6) the P270 enhanced for both stimuli. The occipital P270 enhancement seems to correlate with the decrease in the degree of significance of the task relevant stimuli.
1. Introduction
This study is concerned with the late components of the averaged visual evoked potentials (VEPs) of the human brain associated with the stimulus discrimination situation. In the typical study of the VEP correlates of the stimulus discrimination or comparison, two task relevant stimuli are presented separately in sequence. A subject must retain the memory of the first stimulus until the second one occurs which carries crucial information for the task solution. Chapman and Bragdon (1964), Shelburne (1972, 1973), Thatcher and April (1976) and Thatcher (1977) have studied the VEP correlates of the semantic information processing and demonstrated the amplitude enhancement of the VEP for the second relevant stimulus. Chapman (1973), using a sequence of four letter and number stimuli, has shown that the VEP amplitude (VEP area) was significantly more positve when the stimuli were task relevant than when the same stimuli were task irrelevant. Chapman has measured the VEP amplitudes for ‘perceive-store’ and ‘perceive-compare-solve’ programs, and found that the amplitudes of the vertex VEPs for the ‘perceive-store’ program were significantly different from those for the ‘perceive-compare-solve’ program. The experiment reported in this paper was designed to analyze the VEP behav103
104
S. Kitajima et al. / VEP and problem-sohirq
iors during the task solution process which was separated into four stages: store, retention, solve and post-solve. In this experiment, three identical stimuli were presented repetitively as the first task relevant triplet (intratrial stimulus positions from 1 to 3) in order to separate the memory process into the store (position 1) and retention (positions 2 and 3) stages. This triplet was followed by the second triplet which was composed of another three identical stimuli (intratrial stimulus positions from 4 to 6). The second triplet separated the task completion into the solve (position 4) and post-solve (positions 5 and 6) stages. Tasks were naming the letters or numbers (letter-number hereafter) and matching the checkerboard patterns between two triplets. Under this condition the retained memory was verbal for the letter-number stimulus, while it was non-verbal for the checkerboard stimulus. When identical stimuli are presented repetitively at regular intervals, an initial stimulus produces the VEP and auditory evoked potential (AEP) of the largest amplitude (Ritter, Vaughan and Costa, 1968; Butler, 1973; Kitajima, 1978). Studies on human memory have shown that an initial letter or word of a memory list is most easily remembered (Rumelhart, 1977). A subject in this study could draw a task relevant information from any stimulus in the triplets. However, these data of the initial effects suggest that the subject regards the initial stimuli in the triplets as the most significant for the task solution. In the case of the first triplet, the subject perceives and stores the stimulus in position 1. He must retain this memory until the presentation of the second triplet. During the memory retention he can consolidate the retained memory, if necessary, by means of extracting the sensory information carried by the stimuli in positions 2 and 3. These stimuli may be more useful in consolidating the non-verbal memory and less useful in consolidating the verbal one. In other words, these stimuli may be more significant for the task solution when the memory is non-verbal than when it is verbal. The second triplet measured the effects of the task completion (separated into the solve and post-solve stages) upon the VEP amplitudes. As the short-term memory of the first triplet decays as a function of distance in time from its offset, the task must be solved at the beginning of the second triplet. Therefore, the stimuli in positions 5 and 6 (the post-solve stage) become less significant (or more redundant) for the matching task as well as for the naming one. Hypotheses to be proved in this experiment are: (1) the VEP amplitudes for the letter-number and checkerboard stimuli are maximum at intratrial stimulus positions 1 and 4; (2) the VEP amplitude for the checkerboard stimulus is larger than the amplitude for the letter-number one at positions 2 and 3; and (3) the VEP amplitudes for both stimuli become smaller at positions 5 and 6.
S. Kitajima et al. / VEP and problem-solving
105
2. Methods 2.1. Subjects Subjects were 11 college students, seven males and four females. They ranged in age from 21 to 27, had normal or corrected vision and no history of neurological or visual defects. 2.2. Stimulus presentation Three patterns were used (fig. 1). They were composed of alternating light and dark squares and rectangles on the same-sized checkerboard with 6 X 6 or 12 X 12 cell units. A 6 X 6 cell units complete checkerboard served as a warning stimulus (WS). An incomplete checkerboard composed of six light squares served as a task stimulus. For each column of a 6 X 6 checkerboard, only one cell was light, and the vertical position of the light cell was chosen at random. Another task stimulus was a letter (from captal A to Z, for six subjects) or number (0 and 2 to 9, for five) composed of light cells on a 12 X 12 checkerboard. These task stimuli were matched in total luminance. The stimuli subtended a visual angle of ab’out 3”. Stimulus luminance was 0.05 ml and exposure duration was 300 ms. The stimuli were
WARN ING
ST IMIJLIJS
CHECKERBOARD
LETTER
NUMBER
Fig. 1. Stimulus patterns. A fixation point (not shown) was located at the center terns. Each stimulus was exposed as a light figure against a dark background.
of the pat-
106
S. Kitajima et al. 1 VEP and problem-solving
projected on a translucent screen by a Kodak incandescent projector with a mechanical shutter mounted outside the experiment room. The shutter was activated by an electric pulse, rise and fall time were about 20 ms. There was no shutter noise in the experiment room. Two trials were used: a checkerboard trial and a letter or number trial. The checkerboard and letter trials were used for six subjects, while the checkerboard and number trials for the other subjects. Each trial was composed of seven successive stimuli: the WS, the first triplet consisting of three identical stimuli (intratrial stimulus positions 1,2 and 3) and the second triplet consisting of another three identical stimuli (intratrial stimulus positions 4, 5 and 6). If the frist triplet was composed of either checkerboards, letters, or numbers, the second one was composed of either checkerboards, letters, or numbers, respectively. Fifty checkerboard and 50 letter or number trials were presented in four blocks. About half of 25 trials in the block were the checkerboards and another half were letters or numbers which were arranged at random. The interstimulus interval (ISI) between onsets of the WS and the stimulus in position 1 was 1500 ms. The IS1 between the stimuli in positions 3 and 4 was 1500 ms. The stimuli in each triplet were separated with the IS1 of 1000 ms. The intertrial interval was varied from 20 to 25 s with a mean of 23. The stimulus duration and ISIS were controlled by a preset digital timer. 2.3. Procedure
The subject was seated comfortably in a reclining dentist’s chair and faced a screen in a darkened, sound-proof and electrical shielded room. After becoming fully adapted to the darkness, he was instructed to look binocularly at a faint red fixation point during the stimulus presentation. The centers of the stimuli coincided with the fixation point. The subject was required to perform two tasks. If the stimulus was a checkerboard, he had to make a pattern matching between the first and second triplets and to answer as to whether the triplets were the same or different. If the stimulus was a letter or number, he had to name letters or numbers in the triplets. To avoid myogenic effects on electroencephalographic (EEG) responses, he was asked to answer about 1 s after the termination of the trial. After four practice trials, 100 trials were presented in four blocks. An interval between the blocks was about 3 min, serving as a short rest. During the rest period, a stimulus-mounted tray was changed with another. A few seconds before the presentation of the block, the subject was signalled to look at the fixation point by a brief tone of 1 kHz. After the block presentation, he was signalled to relax by another tone. 2.4. Recording system and data analysis EEG responses were recorded from frontal, temporal, vertex and occipital scalp locations (F3, F4, C3, C4, Cz, 01, and 02 in the international lo-20 system),
S. Kitajima et al. / VEP and problem-solving
107
referenced to linked ears, using chlorided silver disk electrodes. Interelectrode resistances were less than 5 kfi. Eye movements were recorded from electrodes placed on the lateral canthus and the supercilium. EEG and electrooculographic (EOG) responses were amplified by a San-ei 14-channel electroencephalograph (time constant: 0.3 s) and recorded by a TEAC 14-channel FM magnetic tape recorder. EEG responses contaminated by electromyographic (EMG) responses and large eye movements were discarded. Forty EEG responses to the checkerboard trials were averaged using a minicomputer HP-2100 at an analysis time of 3 s (giving a resolution of 7.8 ms/point). Forty EEG responses to the letter or number trials were also averaged in the same way. Averaged EOGs and VEPs were plotted with a Riken X-Y plotter. The peak amplitude was measured with respect to a baseline chosen as the mean voltage over the first 40 ms of the post-stimulus EEG responses. The statistical significance of the amplitude differences between the tasks and across the stimulus positions was evaluated by means of the analysis of variance and the Student’s t-test for paired observations.
3. Results No subjects failed to solve the tasks. In the post-experiment interview, they reported the strategies which they employed to solve the tasks. The stimuli in intratrial stimulus positions 1 and 4 were regarded by all subjects as the most important and looked at most carefully. The matching task was reported as more difficult than the naming. There were no significant differences between the VEP behaviors for the letter and number stimuli. They varied with changes in the stimulus position in the same way. Therefore, these VEP data were joined together in the following analysis. VEP waves recorded from seven electrode loci for a subject are seen in fig. 2. For the VEPs recorded from the centro-frontal five electrodes (F3, F4, C3, C4 and Cz) the most negative peak between 110 and 130 ms, with a mean latency of 120, after the stimulus onset was termed N120, the most positive one between 170 and 200 ms, with a mean latency of 180, was termed P180, and the most positive one between 300 and 460 ms, with a mean latency of 350, was termed P350. For the VEPs recorded from the occipital electrodes (01 and 02), the most negative peak between 160 and 200 ms, with a mean latency of 180, was termed N180, and the most positive one between 250 and 290 ms, with a mean latency of 270, was termed P270. The amplitudes of these peaks recorded from the centrofrontal loci are seen in fig. 3(a), and those recorded from the occipital loci in fig. 3(b). These amplitudes were averaged values across 11 subjects for Cz and 02, while across 10 for C3, C4, and 01, and across nine subjects for F3 and F4, because of EMG and EOG contaminations and poor electrode condition. The N120 amplitudes at the centro-frontal loci were not significantly different either across the stimulus positions or between
’
30-78 ,,.. 8-
1
500 m*ec
POSITION 2
500miec
POSITION
3
LETTER
-NUMBER
4
iO0 inrec MAT&J VJo~‘~50~n*ci~ 8
POSITION
CHECKERBOARD POSITION
5
‘50imw
POSITION 6
I‘ip. 2. VLP waveforms for a subject. tach VIP wils obtained through averaging 40 1 1% responses recorded monopolarly from the centro-frontal (F3. 1’4. C3. C4 and Cz) and occipital (01 and 02) electrodes with reference to the linked ears. The electrode IOCIare wzn on the left and middle ordinates. The VtPs to the letter-number stimulus (the number for this subject) wxe traced with solid lines, whereas the VEPs to the checkerboard stimulus were traced with dotted lines. Positions from 1 to 6 stand for the intratrial stimulus positions. Latencies in msec are seen on the abscissa. The stimuli were presented at 0 ms and terminated at 300 ms. In the case of this subject, the latencies of the centro-frontal N120, P180 and ~350 were about 120, 180 and 300 ms respectively, and those of the occipital NlXO and P270 were about 180 and 270 ms, respectively. The ccntro.frontal PI80 was maximum in amplitude at position 1 and P350 was maxunum in amplitude at posltions 1 and 4. The amplitudes decreased 3c~oss the stimulus positions. The occipital P270 enhanced (more positive) at position 2 in the flr$t triplet for the letter-number stimulus. but not for the clleckerboard stimulus. At position 5 in the second triplet, the P270 enhanced (more pusitwcl for the letter-number and chcckcrboard
MATS;1
POSITION
-
S. Kitajima et al. / VEP and problem-solving Chn
10A’ 6
P350
P180
*+
A I3
N120
109
.*
.
Cm-11
A
‘*
P350
8
If
P180
A
\
B
**
._N 120
01 n=lO P270
A:.10
-
LN
02
B:
_.---
CK
P270
n=ll
4:* B:
8 -
12 .
-4 -
3 .
.
45 .
.
6 -
12
*
*
3
.
45
.
.
6
.
B:*
-6 Fig. 3. The VEP amplitudes measured from the baseline for the letter-number (LN) (traces with solid Lines) and checkerboard (CK) (traces with dashed lines) stimuli. The amplitude values are plotted as a function of the intratrial stimulus positions from 1 to 6 which are seen on the abscissa. Fig. 3(a) shows the amplitudes of the N120, P180 and P350 of the VEPs recorded from the centro-frontal (F3, F4, C3, C4 and Cz) electrodes, and fig. 3(b) shows the amplitudes of the N180 and P270 of the VEPs recorded from the occipital (01 and 02) electrodes. The amplitude values were averaged across 11 subjects for the VEPs from Cz and 02, while across 10 for C3, C4, and 01, and across nine subjects for F3 and F4. The amplitudes in ~.IVare seen on the left ordinate. The capital letters A and B indicate the degree of the significance of the amplitude differences revealed by the analysis of variance. A is for the difference between the letter-number and checkerboard stimuli, and B is for the difference across the stimulus positions (*: p < 0.05; **: p < 0.01). An asterisk between the amplitude values for the letter-number and checkerboard stimuli shows that the student’s t-test revealed significant differences between these amplitudes 0, < 0.05). As seen in fig. 3(a), the amplitudes of the P180 at the centro-frontal loci were maximum at position 1 and decreased across the stimulus positions. The P350 amplitudes were largest at positions 1 and 4, and decreased at the other stimulus positions. At position 4, the P350 to the checkerboard stimulus was larger in amplitude than the P350 to the letter-number stimulus. As seen in fig. 3(b), the occipital N180 amplitudes were maximum at positions 1 and 4, and decreased across the stimulus positions. The P270 amplitudes increased at position 2 for the letter-number stimulus but not for the checkerboard stimulus. The P270 amplitudes at position 5 for the letter-number stimulus were significantly larger than those for the checkerboard stimulus. Though not significant, the P270 for the checkerboard stimulus increased in amplitude across the stimulus positions from 4 to 6. A: 0.10 in 01 indicates p < 0.10.
110
S. Kitajima et al. / VEP and problem-solviq
the letter-number and checkerboard stimuli. The P180 amplitudes at these loci were largest for the stimulus in position 1. The analysis of variance revealed that the decreases of the P180 amplitudes across the stimulus positions were significant: for five loci [F(5, 96) =3.79, p
S. Kitajima et al. / VEP and problem-solving
111
post-solve stages) upon the VEP behaviors were analyzed. The behaviors of the centro-frontal (F3, F4, C3, C4 and Cz) P350 proved the hypotheses that the VEP amplitudes are maximum at intratrial stimulus positions 1 and 4 and become smaller at positions 5 and 6, but did not prove the hypothesis that the VEP amplitude for the checkerboard stimulus is larger than the amplitude for the letter-number one at positions 2 and 3. The behaviors of the occipital P270 did not prove any of them. The enhancement of the centro-frontal P350 was observed for the store and solve stages (at positions 1 and 4). This is consistent with reports of the late positive wave (P300 or P3 in some reports) in relation to the significance of the task relevant stimulus (Smith, Donchin, Cohen and Starr, 1970; Harter and Salmon, 1972; Ritter, Simson and Vaughan, 1972; Wilkinson and Lee, 1972;Picton and Hillyard, 1974; Rohrbaugh, Donchin and Eriksen, 1974; Courchesne, Hillyard and Galambos, 1975; Squires, Donchin, Herning and McCarthy, 1977; reviews NIXmen, 1975; Sabat, 1978), and seem to correspond with the subjects’ reports that they regarded the stimuli in positions 1 and 4 as most significant for the task solution. Chapman (1973) using letter and number stimuli, has reported that the vertex VEP areas (mean amplitudes in his term) during 480 ms after the stimulus onset were significantly different between the store and solve stages. In the present study, the vertex P350 amplitude for the checkerboard stimulus was different between the store and solve stages, but the P350 for the letter-number stimulus was not different between these stages. The amplitude of the human VEP is influenced by the physical and temporal stimulus properties and subjective psychophysiological factors. Many studies on recovery processes of human VEPs have demonstrated that the recovery of the VEP amplitude is dependent upon an interstimulus interval (1%) (Ciganek, 1964; Schwartz and Shagass, 1964; Spehlmann, 1965; Vasconetto, Floris and Morocutti, 1971; Davis, Osterhammel, Wier and Gjerdingen, 1972; Lehtonen, 1973; Kitajima, Morotomi and Kanoh, 1975; Kitajima, 1978). The amplitude decreases of the centro-frontal P180 at positions 2, 3, 5 and 6, the slight increase at position 4, and the amplitude changes of the occipital N180 across the stimulus positions can be explained in terms of the recovery processes. Chapman (1973) has reported that the vertex VEP areas during 225 ms after the stimulus onset decreased as a function of the stimulus positions. An interesting finding in this study was the amplitude increases of the occipital P270 during the retention stage. This VEP enhancement was observed for the letternumber stimulus, bot not for the checkerboard stimulus. Another finding was the amplitude increase of the same P270 during the post-solve stage which.was observed for both stimuli. The physical properties of the stimulus could contribute to the occipital P270 enhancement. MacKay and Jeffreys (1973) Spekreijse, van der Tweel and Zuidema (1973), Jeffreys (1977) and Spekreijse, Estevez and Reits (1977) have demonstrated that different pattern stimuli can produce different amplitudes of the occip-
112
S. Kitujinu et al. / VEP and prohkm-solving
ital VEPs. The differences in the physical properties between the letter-number and checkerboard stimuli may explain the differences between the VEPs to these stimuli, It was observed in this experiment that the occipital P270 alnplitudes for the letter-number stimulus in position 1 (the store stage) were almost the same as those for the checkerboard stimulus in the same position. Therefore, it seems unlikely that the physical properties can explain why the P270 increased significantly in amplitude at position 2 (the retention stage) only for the letter-number stimulus. The amplitude increase of the occipital P270 during the post-solve stage (at positions 5 and 6 in the second triplet), though the P270 enhatl~ement for the checkerboard stimulus was not significant, suggests that the redundancy (a decrease of an amount of necessary information) of the stimulus for the task solution may be one of the factors which contributed to the P270 enhancement. As the short-term memory of the first triplet decays as a function of distance in time from it’s offset, the tasks must be solved at the beginning of the second triplet. The subjects’ reports and the enhancement of the centro-frontal P350 at ~~osition 4 support the hypothesis that the tasks were completed most probably at this stimulus position. After this solve stage, the stimuli in positions 5 and 6 might become redundant for the task solution. It is possible to correlate the occipital P270 enhancement during the post-solve stage with the progress of the task-redundancy at these stimulus positions. In the case of the first triplet, the vertex P350 amplitude and the subjects’ reports seem to support the hypotheses that the matching task was more difficult than the naming one and that during the retention stage the checkerboard stimuli were more significant than the letter-number ones. The enhancement of the occipital P270 seems to correlate with the less-significance or task-redundancy of the letter-number stimulus during the retention stage. There are reports which have demonstrated that the AEP and VW do not habituate to repetitive stimuli if they are significant for the task solution (Sutton, Tueting, Zubin and John, 1967; Courchesne, Hillyard and Courchesne, 1977; reviews Callaway, 1973; Picton, Hillyard and Galambos, 1976). The VEPs to repetitive stimuli should become smaller in amplitude if these stimuli are task-redundant. The VEP enllarl~~ment reported in this paper seems to be inconsistent with the habituation principle. The repetition of the identical task relevant stimuli and the decrease in the degree of significance of these stimuli may contribute to this enhancement. Further analysis of this problem is in progress in our laboratory.
References Butler, R.A. (1973). The cumulative effects of different stimulus repetition rates on the auditory evoked response in man. Electrocncephalography and Clinical Ncurophysiology, 35, 337-345. Callaway, E. (1973). Habituation of averaged evoked potentials in man. In: Pecke, H.V.S. and
S. Kitajima et al. / VEP and problem-solving
113
Herz, M.J. (Eds). Habituation Vol. II. Physiological Substrates. Academic Press: New York, 153-174. Chapman, R.M. and Bragdon, H.R. (1964). Evoked responses to numerical and non-numerical visual stimuli while problem solving. Nature (London), 203, 1155-1157. Chapman, R.M. (1973). Evoked potentials of the brain related to thinking. In: McGuigan, F.J. and Schoonover, R.A. (Eds). The Psychophysiology of Thinking. Academic Press: New York, 699108. Ciganek, L. (1964). Excitability cycle of the visual cortex in man. Annals of New York Academy of Science, 112,241-253. Courchesne, E., Hillyard, S.A. and Courchesne, R.Y. (1977). P3 waves to the discrimination of targets in homogeneous and heterogeneous stimulus sequences. Psychophysiology, 14,590597. Courchcsne, E., Hillyard, S.A. and Galambos, R. (1975). Stimulus novelty, task relevance and the visual evoked potential in man. Electroencephalography and Clinical Neurophysiology, 39,131I143. Davis, H., Osterhammel, P.A., Wier, C.C. and Gjerdingen, D.B. (1972). Slow vertex potentials: Interactions among auditory, tactile, electric and visual stimuli. Electroencephalography and Clinical Neurophysiology, 33,537-545. Harter, M.R. and Salmon, L.E. (1972). Intra-modality selective attention and evoked cortical potentials to randomly presented patterns. Electroencephalography and Clinical Neurophysiology, 32,605-613. Jeffreys, D. (1977). The physiological significance of pattern visual evoked potentials. In: Desmedt, J.E. (Ed.). Visual Evoked Potentials in Man: New Development. Clarendon Press: Oxford, 134-167. Kitajima, S. (1978). The cumulative inhibitory effect of repetitively flashed stimuli on the recovery process of the human visual evoked potential to a test stimulus. Electroencephalography and Clinical Neurophysiology, 44, 364-372. Kitajima, S., Morotomi, T. and Kanoh, M. (1975). Enhancement of evoked responses to brief flashes after offset of preexposed light stimulation: A critical moment. Vision Research, 15, 1213-1216. Lehtonen, J.B. (1973). Functional differentiation between late components of visual evoked potentials recorded at occiput and vertex: Effect of stimulus interval and contour. Electroencephalography and Clinical Neurophysiology, 35,75-82. MacKay, D.M. and Jeffreys, D.A. (1973). Visually evoked potentials and visual perception in man. In: Autrum, H., Jung, R., Loewenstein, W.R., MacKay, D.M. and Teuber, H.L. (Eds). Handbook of Sensory Physiology Vol. VII/3B. Springer-Verlag: Berlin, 647-678. N%&ren, R. (1975). Selective attention and evoked potentials in humans - A critical review. Biological Psychology, 2,237-307. Picton, T.W. and Hillyard, S.A. (1974). Human auditory evoked potentials. II. Effects of attention. Electroencephalography and Clinical Neurophysiology, 36, 191-200. Picton, T.W., Hillyard, S.A. and Galambos, R. (1976). Habituation and attention in the auditory system. In: Keidel, W.D. and Ncff, W.D. (Eds). Handbook of Sensory Physiology Vol. 5. Springer-Verlag: Berlin, 343-387. Ritter, W., Simson, R. and Vaughan, H.G. Jr. (1972). Association cortex potentials and reaction time in auditory discrimination. Electroencephalography and Clinical Neurophysiology, 33, 547-555. Ritter, W., Vaughan, H.G. Jr. and Costa, L.D. (1968). Orienting and habituation to auditory stimuli: A study of short term changes in average evoked responses. Electroencephalography and Clinical Neurophysiology, 25,550-556. Rohrbaugh, J.W., Donchin, E. and Eriksen, C.W. (1974). Decision making and P300 component of the cortical evoked response. Perception and Psychophysics, 15, 368-374.
114
S. Kitajirna et al. / VEP and problem-solving
Rumelhart, D.E. (1977). Introduction to Human Information Processing. John Wiley, New York. Sabat, S.R. (1978). Selective attention, cortical evoked responses, and brain function in human subjects: A critical review and theory. Biological Psychology, 7, 175~-201. Schwartz, M. and Shagass, C. (1964). Recovery functions of human somatoscnsory and visual evoked potentials. Annals of New York Academy of Science, 112, 510-525. Shelburne, S.A. Jr. (1972). Visual evoked responses to word and nonsense syllable stimuli. Electroencephalography and Clinical Neurophysiology, 32, 17--25. Shelburne, S.A. Jr. (1973). Visual evoked responses to language stimuli in normal children. Electroencephalography and Clinical Neurophysiology, 34, 135-143. Smith, D.B.D., Donchin, I:., Cohen, L. and Starr, A. (1970). Auditory averaged evoked potcntials in man during selcctivc binaural listening. Electroencephalography and Clinical Neurophysiology, 28, 146-152. Spehlmann, R. (1965). The averaged electrical rcsponscs to diffuse and to pattcrncd light in human. Electroencephalography and Clinical Neurophysiology, 19, 560 --569. Spckreijse, H., Estevez, 0. and Rcits, D. (1977). Visual evoked potentials and the physiological analysis of visual processes in man. In: Dcsmedt, J.11. (Ed.). Visual Evoked Potentials in Man: New Dcvclopments. Clarendon Press: Oxford, 16-89. Th. (1973). Contrast evoked responses in Spekreijse, H., van der Twecl, L.H. and Zuidcma, man. Vision Research, 13, 1577-1601. Squires, K.C., Donchin, E., Her&g, R.I. and McCarthy, G. (1977). On the influence of task relevance and stimulus probability on event-related-potential components. Elcctrocncephalography and Clinical Neurophysiology, 42, l-14. Sutton, S., Tucting, P., Zubin, .I. and John, E.R. (1967). Information delivery and the sensory evoked potential. Scicncc, 155, 1436-1439. Thatcher, R.W. (1977). Evoked-potential correlates of delayed letter-matching. Behavioral Biology, 19, l-23. Thatcher, R.W. and April, R.S. (1976). Evoked potential correlates of semantic information processing in normals and aphasics. In: Richer, R.W. (Ed.). The Neuropsychology of Language. Plenum Press: New York, 95-124. Vasconetto, C., Floris, V. and Morocutti, C. (1971). Visual evoked responses in normal and psychiatric subjects. Electrocncephalography and Clinical Neurophysiology, 3 1, 77 --83. Wilkinson, R.T. and Lee, M.V. (1972). Auditory evoked potentials and selective attention. k:lectroencephalography and Clinical Neurophysiology, 3 3,4 11-4 18.