вход по аккаунту


Differences in reaction times and average evoked potentials as a function of direct and indirect neural pathways.

код для вставкиСкачать
Differences in Reaction Times
and Average Evoked Potentials as a
Function of Direct and Indirect
Neural Pathways
Alexa Ledlow, PhD, James M. Swanson, PhD, and Marcel Kinsbourne, M D
Average evoked potentials and manual response latencies were collected during a simple detection task in which
brief visual stimuli were presented to the left and right visual fields. Latencies generated by the ipsilateral stimulushand combinations were shorter than contralateral combinations only under certain conditions, impugning the
hypothesis that the reaction time difference reflects interhemispheric transfer time. Certain evoked potential
components recorded conualateral to the stimulus occurred earlier than their ipsilateral counterparts, but whether
this difference can be interpreted as representing interhemispheric transfer time is also questioned.
Ledlow A, Swanson JM, Kinsbourne M: Differences in reaction ames and average evoked potentials as a
function of direct and indirect neural pathways. Ann Neurol 3:525-530, 1978
Both the visual pathways from the eye to the brain
and the motor pathways from the brain to the hand
are contralateral in humans. Based on this organization, theories have been proposed suggesting that
when a visual stimulus is presented to one side of the
vertical meridian and requires a reponse movement
by the ipsilateral hand, then all processing can be
handled within one cerebral hemisphere. If, however, a response is to be made by the other hand,
according to this theory, information must flow via
the corpus callosum to the hemisphere ipsilateral to
the stimulus before the response can be initiated.
Poffenberger [16] first made this argument and proceeded to measure the visual-manual response latencies of each hand of a subject when visual stimuli
were presented to either the right or the left visual
field. By subtracting reaction times (RTs) obtained
from the ipsilateral hand-visual field conditions from
RTs obtained from the contralateral hand-visual field
conditions, he attempted to calculate interhemispheric transfer time (IHTT).His estimate of IHTT
was 4 msec, and this estimate is consistent with predictions, based on known neural conduction time and
synaptic delay time, of how long is required for information to cross the short interhemispheric distance.
Interpretation of the contralateral-ipsilateral RT
differences as IHTT has been challenged on the
grounds that in various experiments, widely different
estimates of IHTT can be obtained [20]. A review of
IHTT experiments revealed that the RT advantage
for direct projection to the hemisphere controlling
the response was larger when the stimulus location
was unknown to the subject 12, 81 than when it was
completely predictable [ 1, 161. Experiments using
vocal responses that must be programmed by the left
hemisphere provide an analogous contrast. Using
known (predictable) stimulus locations, Moscovitch
and Catlin [15] found a 10 msec rlght visual field
advantage, but with intermixed (and thus unpredictable) stimulus locations, Filbey and Gazzaniga [5]
found the right field advantage to be three times as
large (33 msec). Carmon et al C31 found that known
versus unknown stimulus location had the same effects on RT and IHTT. In addition to these intentional measurements of I H I T , the same effect of
predictable and unpredictable locations may be deduced from data not designed specifically to study
IHTT or hemispheric specialization [ 181.
Two obvious explanations for the discrepancies in
IHTT estimates exist. First is the possibility that fixation will be biased when subjects know where the
stimulus will appear. If this occurs, IHTT measurements will be anifactually low. Second is the possibility that the RT difference reflects output factors;
when the location is constant, the effects of
From the Neuropsychology Research Unit, The Hospital for Sick
Children, Toronto,Ont, Canada.
Address reprint requests to Dr Ledlow, Division of Behavioral
Science, University of Wisconsin-Parkside, Kenosha, W1 53141.
Accepted for publication Dec 30, 1977.
0364-5134178/0003-0611$01.25 @ 1978 by Alexa Ledlow
stimulus-response compatibility are reduced. If the
first possibility is the case, then t h e logic of IHTT
m e a s u r e m e n t by RTs holds, b u t t h e precision of t h e
technique is questionable. If the second explanation
is valid, IHTT m e a s u r e m e n t by RT methodologies
may be discredited.
The e x p e r i m e n t described here was designed to
use manual RTs a n d e v o k e d brain responses t o
explore the suitability of the IHTT paradigm. Special
care was t a k e n to e n s u r e that t h e subject was focusing
on t h e midline fixation cue p r i o r t o stimulus delivery
and that no e y e movements occurred d u r i n g a trial.
I n order t o obtain responses that depended only on
direct versus indirect (transcallosal) reception of the
visual system (uncontaminated by s u b s e q u e n t o u t p u t
activities), evoked potentials were recorded from the
visual p r o j e c t i o n area of each hemisphere.
Eight normal young adults served as subjects. Four were
laboratory personnel who were unaware of the experimental hypothesis and 4 were student recruits who were paid
for their participation. All were right-handed.
The following five dichotomous within-subject variables
were incorporated into the design of this experiment.
1. Left and right visual field. The stimulus, which was a
square subtending 1 degree of visual angle, was displayed
2.5 degrees to the left or right of midline.
2. Targets and nontargets. Half o f the stimulus squares
were inscribed with an X; half were empty. For half the
subjects, the squares with an X were designated targets
requiring that a manual button be pressed while the empty
squares required the subject to withold response. The opposite designation was used for the other subjects.
3. Left and right response hand. Each sequence of trials
was run twice, once with the subject using his right index
finger to press a centrally located button for responding to
target stimuli, and once using the left index finger. The
order was reversed for half the subjects.
4. Known and unknown stimulus location. Trial sequences contained only stimuli in one visual field (known
location) or left and right visual field stimuli randomly
mixed (unknown location). Half the subjects received the
known location sequences first and half, the unknown sequences.
5. Left and right hemisphere. For analyses of the average
evoked potentials (AEPs),the additional factor of left and
right hemisphere applies, since electroencephalographic
recordings were made from both hemispheres during all
Each subject participated in a familiarization and practice
After electrodes had been applied for monitoring the left
and right hemispheres and vertical and horizontal eye
526 Annals of Neurology Vol 3 No 6 June 1978
movements, the subject was seated in front of a C R T
screen that displayed all stimuli. Head position was
stabilized in a Narco-bio chin rest and head-holdinR fixture.
Subjects were instructed to respond as quickly as possible
to targets without making errors by responding to nontargets. Before each sequence they were told which hand to
use for responding and whether the squares would appear
all on the left, all on the right, or randomly mixed between
the left and right side of the display.
At the beginning of each trial, a centrally located square
subtending 0.3 degree of visual angle and a dot that was
randomly located within a l o d e g r e e radius of the square
appeared o n the screen. T h e subject moved the dot into
the square by manipulating a control stick with the hand
not being used for pressing the button. T h e purpose of
requiring the subject to center the dot was to force fixation
within 0.3 degree before each trial began. Since the subject's head was in a fixed position, it was possible to determine the position of the eyes for use as a fixation baseline
on each trial to allow comparison of subsequent contaminating movements of the eyes.
After the dot was centered the square disappeared, leaving the dot in a central position as the fixation cue. One
second later the stimulus appeared for 40 msec duration.
The fixation dot remained for 1 second after stimulus onset. T h e screen was then blank for 1 second before commencement of the next trial. An auditory signal was delivered following errors, and the trial was repeated later in the
same sequence. Sequences with all stimuli in one visual
field were composed of 64 stimuli; those with stimuli in
both fields, 128 stimuli (plus the repeated trials).
Subjects were strongly admonished to keep their eyes on
the central fixation cue as long as it was visible and to
refrain from making any movement during that period.
They were encouraged to blink, cough, or make any other
movements only before they began to center the dot on
each trial. Recordings were examined after the practice session before the experiment was continued. Several subjects
were found to be making saccades toward the stimulus
about 200 msec after stimulus onset. These subjects were
carefully reinstructed and, though they seemed unaware
they had been making these movements, were able to inhibit them in subsequent sessions.
Recording Procedures and Equipment
Cerebral potentials were recorded using Beckman biopotential electrodes placed over the occipital areas (between
0, and P3 o n the left and 0, and P, on the right). Both
active electrodes were referred to the Linked mastoids. T h e
electrooculogram was recorded between infraorbital and
supraorbital ridges of the right eye and between the outer
canthi. Grass P122 amplifiers (DC setting) were used to
measure the AEP signals. Signals were transformed on line
by an analog-todigital convener and stored o n electromagnetic discs for later processing. Stimulus presentation and collection as well as storage and processing of data
were controlled by a DECLAB 11/40 computer system.
Sampling of the horizontal eye channel began when the
target cue was centered. The sampling epoch for the other
channels began 40 msec prior to stimulus onset and continued for 800 msec. The sampling rate was 250 Hz.
All four channels of electrophysiological data from each
trial were viewed simultaneously on the display screen. All
data were deleted from trials in which eye movements,
blinks, or other artifacts were detectable. The average
number of deletions per subject and condition was 1.6;the
greatest number was 6.
Addit io nul Conditions
After 4 subjects had participated in the experiment it was
observed that, contrary to expectations, responses made
with the hand ipsilateral to the stimulus were no faster than
those with the contralateral hand. Some subjects showed
small RT differences in the opposite direction, even for the
sequences in which stimuli appeared randomly in both visual fields. If stimulus-response compatibility is an important factor in this paradigm, then the position of the response hand could be crucial. The remaining 4 subjects
therefore participated in a replication of the experiment in
which the response button was located 15 cm lateral to the
midline toward the side of the response hand. Two subjects
responded using the central button first, and 2 used lateral
buttons first.
The last 4 subjects also participated in a cohtrol condition in which no response was required. This condition
would permit determination of whether any effects of direct or indirect reception on the AEPs depended o n the
rcquirement to attend to and discriminate the stimuli. The
subject went through the fixation procedure as usual on
each trial and simply watched the stimulus presentation
without making any response. This “no-response’’condition was always run as the final session of the experiment.
Averaged over subjects and conditions, 0.47 error
was committed per 32 correct responses. Analyses of
variance for the error data revealed no significant effects.
Reaction Times
An analysis of variance was computed for the first 4
subjects’ mean RTs to target stimuh using the central
response button. The factors were Visual Field, Response Hand, and Known versus Unknown Stimulus
Location. As initial inspection of the data had suggested, no significant effects were found. RTs for the
ipsilateral hand-visual field combination were only 2
msec less than for the contralateral combination.
Another analysis was performed o n the complete
RT data for the second 4 subjects with Button Locdtion as an additional factor. The knowledge of location factor was significant (F(1,3) = 13.01,p C 0.05),
with stimuli in known locations producing faster responses, by 14 msec, than those in unknown locations. The three-way interaction of Visual Field X
Response Hand x Button Location was significant
(F(1,3) =30.00,p C 0.05), and the four-way interaction of these factors with Knowledge of Stimulus Lo-
0 ipriloterol
Fig I . Mebn RTs for ipsilateralandcontra~terdlcombinations
of response hand and stimulusfield f o r each catadition of central
(8.subjects) or lateral t4 subjects) position of response button and
certain or uncertain stimulus location.
cation approached significance (F(1,3) = 9.26, p =
0.056). Separate analyses of simple effects were
therefore performed to determine the different RT
patterns based o n data from the two response locations (central and lateral). In the analysis of data from
the lateral button condition, a significant three-way
interaction of Visual Field x Response Hand x
Knowledge of Stimulus Location (P( 1,3) = 11.69,p C
0.05) and a significant Visual Field x Response
Hand interaction (F(1,S) = 41.50,p < 0.01)were
obtained, indicatjpg that the I H T T estimate (the
contralateral-ipsitateral difference reflected by the
Visual Field X Response Hand interaction) was significantly greater when the stimulus location was unknown than when it was known. In analysis of data
from the same subjects using the centrally positioned
button, no results were significant (Visual Field x
Response Hand x Knowledge of Stimulus Location:
F(1,3) = 0.24, p = 0.66; Visual Field X Response
Hand: E(1,3) = 0.12,) = 0.75). Figure 1 shows the
mean RTs for the response hands contralateral and
ipsilateral to the stimulus field for each condiaon.
Eye M0vernent.r
The horizontal eye movement averages revealed only
small, unsystematic movements that could not be telated to the side of stimulation. The records appeared
the same for known and unknown locations, indicating that the attempts to control eye movements were
Averaged Evoked Potentials
Negative peaks were commonly located at approximately 70 and 170 msec following stimulus onset,
Ledlow, Swanson, and Kinsbourne: Comparison of RT and AEP Differences 527
Table 1 , Amplitudes and Lutenrics of AEPs from Each Hemisphm for Stimuli in Each Visual Field
Amplitudes (pv)
Left Hem.
Latencies (msec)
Right Hem.
Left Hem.
Right Hem.
Central response
Lateral response
No response
'p < 0.01.
bp < 0.05.
LVF = left visual field; RVF
= right visual field.
and positive peaks at 130 and 240 msec and after 300
msec. The amplitudes relative to mean voltage of the
initial 40 msec of the sampling epoch and the latency
of these five peaks (N70,P130, N170, P240, and
P300) were used as dependent variables. Separate
analyses of variance were calculated for each of the
dependent measurements from the central response,
lateral response, and no-response sequences. Five
factors were used in each analysis (see Design). For
the no-response analyses, the response hand was the
one not manipulating the joy stick in that sequence,
and targets were the stimuli to which the subject had
responded in his or her other sessions. N o P300
components were apparent in the AEPs from noresponse sequences; nevertheless, as a control procedure for the purpose of analysis, the most extreme
positive point following P240 was designated P300.
In the analysis of two of the early AEP peaks (P130
and N170),significant Visual Field x Hemisphere
interactions were obtained. The data and significance
levels are shown in Table 1. This interaction did not
approach significance for the other AEP peaks "70,
P240, and P300). The significant Visual Field x
Hemisphere effect indicates that the dxect versus
indirect routes of visual input produced differences
in the cortical response as measured by the AEP. The
effects on the AEPs of direct versus indirect projection routes for the three response conditions are
shown graphically in Figure 2.
In contrast to the RT data, the Knowledge of Location factor was not significant, nor did it interact with
any other factor. There were significant target versus
nontarget differences in P240 and P300 in several
analyses, as shown in Table 2, but no significant differences for this factor occurred in the tests on earlier
components of the AEP.
Annals of Neurology Vol 3 No 6 June 1978
5~ vohr
100 150 200 250 300 350 400
FiK 2. CompositeAEPs based on average peak latencies and amplitudcr. Direct AEPs wen recordedfrom the hemisphere contrakateral to the side of stimulus presentation, indirect AEPs
from the ipsikateral hemisphm.
In typical evoked potenaal studies of selective attention, only target stimuli elicit P300 components [7,
191. However, in a reaction-time regimen it is common for all stimuli to elicit a large P300 [4]. The
Table 2. AEP Ampiituk and Lutenry Diffnnrres for Targets and Nontargets
Amplitudes (pv)
Central response
Iaceral response
N o response
‘p < 0.01.
Iatencies (msec)
Nontarge t
Nontarge t
24 1
24 3
bp € 0.05.
effect on P300 of target and nontarget stimuli observed in this experiment was similar to that reported
by Posner et al [17]. In their experiments, stimuli
that matched or that the subject was instructed to
count produced a larger and earlier P300.
The RT and AEP results of this experiment provide an interesting contrast. Whereas under circumstances of active processing, indirect projection results in evoked potential waveform alterations which
include delay of certain peaks, indirect versus direct
pathways are not a critical factor in producing RT
First, consider the RT findings. When a central response was made, ipsilateral responses had no RT
advantage over contralateral ones. Yet by slight alterations in the position of the hand, we were able to
produce RT differences that might erroneously be
labeled IHTT. However, this pattern emerged only
when stimulus location was unknown.
IHlT should not vary with the subject’s changing
expectation of the stimulus locations (although one
might argue, as did Gazzaniga [16], that I H T T will
increase as more information is transmitted). There is
no simple explanation for a significant alteration in
I H l T with a change in the position of the hand. But
such a change occurred in this experiment, suggesting that other factors affect RT and contaminate the
IHTT estimate. In retrospect, this result is not surprising; application of Poffenberger logic to a
stimulus-response compatibility study by Wallace [2 13
in which the arms were crossed actually yielded a
negative IHTT, estimated by subtracting RTs resulting from direct neural pathways from those due to
indirect pathways. Thus, in that experiment, response
latency effects were due to the spatial correspondence
of the stimulus and the location of the response, not to
the anatomical relations.
If the thesis that IHTT can be found by Subtracting
RTs must be rejected, what is the explanation for
longer latencies for indirect pathways under some
circumstances? The attentional or orientational
model proposed by Kinsbourne [9- 111 provides a
mechanism to explain the effects of both location
expectancy and hand position.
It is evolutionarily advantageous for stimulation
from one side of space to elicit a synergistic orienting
response (OR) toward the source of stimulation. I n
organisms with crossed neural organization, the OR
is mediated by the contralateral hemisphere. Responses compatible with the OR of the stimulated
hemisphere will be favored. In the present experiment, when the subject was fully prepared for the
stimulus to appear in a set location, no O R was elicited. But when an OR was precipitated by lateral
stimulation in an unpredictable fashion, the responses most compatible with the direction of the
OR were most benefited.
Consider now the electrophysiological results. In
two experimental conditions in which the subjects’
task required them to attend to the stimulus in order
to determine when to make a response, the two middle components of the AEP occurred almost 20 msec
sooner if the stimulus appeared contralateral rather
than ipsilateral to the recording site. Amplitude effects accompanied the delay of P130 and N 1 7 0 .
P130 was over 2 p v larger when reception was indirect. An opposite effect of comparable size occurred
for N 1 7 0 . However, the P 1 3 0 - N l 7 0 amplitudes for
direct and indirect reception are comparable, so that
the baseline amplitude differences at N 1 7 0 probably
reflect the degree of positive bias that occurred earlier. These AEP differences in direct and indirect
projection closely resemble those found in a study of
hemispheric specialization for name and physical
matching using left and right visual field pairs [13].
The lack of s i d c a n t interactions involving the
visual field and hemisphere variables in the analysis
of AEPs from conditions in which the subject
passively observed the stimuli in no-response conditions complicates this interpretation of the data. Ap-
Ledlow, Swanson, and Kinsbourne: Comparison of RT and AEP Differences
parently the direct-indirect differences are dependent on attention or active processing. A similar
effect has been reported for the auditory modality by
Mononen and Seitz [14]. Ipsilateral AEP peaks were
delayed when the subject had to locate clicks within
sentences. But when the clicks were passively received, there was no contralateral over ipsilateral advantage.
It is particularly important to recall, in attempting
to interpret these differences, that the effects of direct versus indirect projection on the AEPs were the
same irrespective of these factors (namely, knowledge of location and response hand position) which
produced changes in the RT differences for direct
and indirect pathways. If analysis of RT measures
cannot capture the elusive IH'IT, perhaps AEP peak
latency measures, which eliminate the added noise
and variability of subsequent response-related factors
of processing, do reflect IHTT.
If I H I T were reflected in indirectly produced
AEPs in a stmghtforward way, the expected pattern
would be AEPs identical to those produced by direct
projection but shifted in time by a constant time
interval. The directly and indirectly derived AEPs in
this experiment did not show such a pattern. Until
120 msec from stimulus onset, the AEPs from the
two hemispheres were no different, suggesting that
activity had been initiated in both the directly and
indirectly stimulated visual areas and progressed
roughly in parallel (within the measurement limitations of the procedures of this study). Interhemispheric transfer must have occurred and failed to be
At about 120 msec the AEPs diverged; the directly
stimulated hemisphere evidenced net negative activity, while the indirectly evoked waveform continued
in the positive direction for almost 20 msec. Transcallosal communication is between association areas of
the cortex. The onset of the N170 wave in this task
may have depended on relatively high level output
from the association areas of the directly stimulated
hemisphere. This was available 20 msec earlier when
interhemispheric transfer was not involved. An alternative and more conservative speculation is that the
differences in AEP during the 120 to 190 msec poststimulus interval were related to slower, less efficient, or in some way different processing by the
transcommissurally stimulated hemisphere.
In any case, the differences are not preserved in
the late positive components. These components,
like manual responses, may be dependent on prior
integration of processing from both hemispheres,
and thus always include IH'IT components. Unlike
the RT measures, however, P300 was not sensitive to
stimulus-response compatibility factors. This supports the orientational interpretation given earlier for
the RT data, since a recent report by Kutas, McCar530 Annals of Neurology Vol 3 No 6 June 1978
thy, and Donchin [I21 has shown that P300 latency is
uninfluenced by response selection.
1. Berlucchi G, Heron W, Hyman R, et al: Simple reaction times
of ipsilateral and contralateral hand to lateralized visual stim-
uli. Brain 94:419-430. 1971
2. Bradshaw JL., Perriment A D Lateraliry effects and choice
reaction time in a unimanual mefinger cask. Percept
Psychophys 7:185-188, 1970
3. Carmon R, Nachshon I, Iseroff A, et al: Visual field differences in reaction times to Hebrew letters. Psychonom Sci
4. Donchin E, Kubovy M, Kutas M, et al: Graded changes in
evoked response (P300)amplitude as a function of cognitive
activity. Percept Psychophys 14319-324, 1973
5. Filbey RA, Gazzaniga MS: Splimag the normal brain with
reaction time. Psychonom Sci 17:335-336. 1969
6. Gazzaniga MS: Reply to McKeever and Huling. Psychonom
Sci 22:222-223. 1971
7. Hillyard S, Hink R, Schwent V, et al: Electrical signs of selectiveattention in thehuman brain. Science 182:177-179,1973
8. Jeeves MA: Psychological studies of three of c o n e n i t d
agenesis of the corpus callosum, in Enlinger EG (ed):Functions of the Corpus Callosum. London, Churchill, 1965,pp
9. Kinsbourne M: The cerebral basis of lateral asymmetries in
attention. Acta Psychol 33:193-201, 1970
10. Kinsbourne M: The control of attention by interaction between the cerebral hemispheres, in Kornblum S (ed):Aaention and Performance IV.New York. Academic Press, 1973.
p p 239-256
11. Kinsbourne M: Mechanisms of hemispheric interaction in
man, in Kinsbourne M, Smith WL (eds): Hemispheric Disconnection and Cerebral Function. Springfield, IL, Charles C
T h o u , 1974,pp 260-285
12. Kutas M. McCarthy G, Donchin E: Augmenting mental
chronometry: the P300 as a measure of stimulus evaluation
time. Science 197:792-795, 1977
13. Ledlow A, Swanson J. Kinsbourne M: Reaction times and
evoked potentials as indicators of hemispheric differences for
laterally presented name and physical matches. J Exp Psychol
[Human Percept] (in press)
14. Mononen IJ, Seiu MR: An AER analysis of conualateral advantage in the transmission of auditory information. Neuropsychologia 15:165-173, 1977
15. Moscovitch M,Catlin J: Interhemispheric transmission of information: measurement in normal man. Psychonom Sci
16. Poffenberger AT: Reaction rime to retinal stimulation with
special reference to time lost in conduction through nerve
centers. Arch Psychol 3:l-73. 1912
17. Posner MI,Klein R, Summen J. et al: O n selection of signals.
Mem C o p 1:2-12. 1973
18. Posner M1. Nissen MJ,Ogden W Attending to a position in
space. Presented at the Psychonomic Society Meeting, Denver, CO, November 1975
19. Schwent VL, Hillyard SA, Galambos R: Selective attention
and the auditory vertex potential: I. Effects of stimulus deliverg rate. Elecuwncephalogr Clin Neurophysiol40:604-614.
20. Swanson J. Ledlow A. Kinsbourne M: Lateral asymmetries
revealed by simple reaction time, in Kinsbourne M (ed): T h e
Asymmetrical Function of the Brain. New York, Cambridge
University Press, 1978
21. Wallace RJ: S-Rcompatibility and the idea of a response code.
J Exp Psychol 88:354-360,1971
Без категории
Размер файла
548 Кб
potential, times, reaction, averages, differences, neural, direct, indirect, function, pathways, evoked
Пожаловаться на содержимое документа