Impaired perception of temporal order in auditory extinction

Transcription

1 Neuropsychologia 40 (2002) Impaired perception of temporal order in auditory extinction Hans-Otto Karnath a,, Ulrike Zimmer a, Jörg Lewald b,c a Department of Cognitive Neurology, University of Tübingen, Hoppe-Seyler-Strasse 3, D Tübingen, Germany b Institut für Arbeitsphysiologie an der Universität Dortmund, Dortmund, Germany c Fakultät für Psychologie, Ruhr-Universität, Bochum, Germany Received 16 January 2002; received in revised form 10 April 2002; accepted 15 April 2002 Abstract It has been proposed that patients with extinction show a chronic bias of spatial attention towards the ipsilesional side. In this case, the law of prior entry predicts that ipsilesional events should be perceived earlier than physically synchronous contralesional stimuli. In line with this prediction, previous studies have revealed substantial delays of awareness for contralesional visual and tactile events in patients with visual and with tactile extinction. The present study provides evidence that a prior entry bias also occurs in the auditory modality. Patients with auditory extinction perceived two acoustic events (one presented to the left ear, the other to the right ear) as being simultaneous when the contralesional sound was leading by 270 ms. The magnitude of this asynchrony was quite similar to that measured previously in the visual modality. Thus, the pathological delay of awareness for contralesional events may be independent of the sensory modality of the stimuli Elsevier Science Ltd. All rights reserved. Keywords: Extinction; Neglect; Temporal order perception; Hearing; Parietal; Brain-damage; Human 1. Introduction Patients with unilateral left or right brain-damage (RBD) may show the phenomenon of contralesional extinction. These patients respond appropriately to visual, auditory, or tactile stimuli on both the ipsilesional and the contralesional side as long as the stimuli are presented in isolation. However, when two stimuli are presented simultaneously, the more contralesionally located stimulus goes undetected (at least to a certain degree). Several observations favour the assumption that spatial attention is chronically biased towards the ipsilesional side in these patients. For example, awareness of stimuli presented on the contralesional side can be improved by instructing extinction patients to ignore any concurrent ipsilesional events [7]. Further, ipsilesional visual events are perceived earlier than physically synchronous contralesional visual stimuli. In a visual temporal order judgement task, Rorden et al. [11] and Baylis et al. [1] investigated patients with left-sided or with right-sided visual extinction. The patients judged the order of onset of two rectangular bars, presented to the left and right of a central fixation point. Both stimuli were presented either simultaneously or with the left or the right stimulus leading in time. The stimulus-onset asynchrony (SOA) at which the Corresponding author. Fax: address: karnath@uni-tuebingen.de (H.-O. Karnath). patients with extinction perceived the two events as being simultaneous (the point of subjective simultaneity) was obtained when the contralesional bar preceded the ipsilesional by over 200 ms. The substantial delay of awareness for contralesional visual events suggested a severe bias in the time-course of visual awareness to the ipsilesional side. A prior entry bias was also observed in the tactile modality. Birch et al. [2] investigated 19 RBD patients with tactile extinction. With tactile stimulation of both forearms, the authors found that the number of extinguished contralesional events significantly decreased when the contralesional side was stimulated prior to the ipsilesional side. Recently, Mattingley examined a further patient with tactile extinction following a right hemisphere stroke [9]. While detection of single vibrotactile stimuli on either hand was near ceiling, there was a severe decrement in detection of left targets when two stimuli were presented. This impairment for left targets occurred over a broad range of SOAs indicating that the occurrence of a right-sided competitor interfered with detection of left targets even when the two events were asynchronous. Since the experimental designs used in both these studies on tactile extinction did not allow to exactly determine the point of subjective simultaneity, a direct comparison with the 200 ms delay measured for contralesional visual events in patients with visual extinction [1,11] is not possible /02/$ see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S (02)

2 1978 H.-O. Karnath et al. / Neuropsychologia 40 (2002) The present study aimed to investigate whether or not a prior entry bias occurs also in auditory extinction, thus affecting the time-course of awareness for auditory events. For direct comparison with the previous results obtained in the visual modality, we used an experimental design similar to that employed by Rorden et al. [11]. In particular, we determined the point of subjective binaural simultaneity for acoustic stimuli, i.e. that time difference between two sound stimuli presented to the left and the right ear, at which the sounds were perceived as being simultaneous. 2. Methods 2.1. Subjects Patients with acute unilateral brain damage, consecutively admitted to the Department of Neurology at the University of Tübingen were screened for auditory extinction by (i) clinical bedside testing and (ii) presentation of standardized sound stimuli via headphones in a sound-proof room. Patients were included if they had circumscribed unilateral brain lesions due to ischemic stroke or hemorrhage demonstrated by magnetic resonance imaging (MRI) or computed tomography (CT). Patients with diffuse or bilateral brain lesions, patients with tumors, as well as patients in whom MRI or CT scans revealed no obvious lesion were excluded. Also not included in the main experiment were three subjects with unilateral hearing loss (see Section 2.3). We found four RBD patients who exhibited auditory extinction. They were compared with a group of five RBD patients without extinction from the same recruitment sample. Five healthy subjects served as additional controls. All 14 subjects gave their informed consent to participate in the study which has been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. Table 1 gives an overview of all subjects included. The neuropsychological test results are presented together with the demographic and clinical data of the subjects Assessment for auditory extinction In an initial test, all patients were clinically examined at the bedside by rustling with small pieces of paper near the patients ears (with the patients eyes closed). Ten of these sound stimuli were presented unilaterally at the left ear, 10 stimuli at the right ear, and 10 stimuli simultaneously to both ears in a quasi-random order. For more elaborate investigation, we used standardized experimental conditions. Subjects sat on a chair in a dark and sound-proof room. The head was stabilized in line with the trunk by a chin rest. Sound stimuli consisted of band-pass filtered noise (cutoff frequencies 0.6 and 5 khz; duration 500 ms; 70 db re 20 Pa) and were presented either monaurally or binaurally via headphones (Sennheiser HD580). For binaural stimulation, we used incoherent noise, i.e. waveforms provided by the two stereo channels were independent. This was done to prevent binaural fusion of the sounds presented to the left and right ear into one unified percept, as occurs with coherent headphone stimuli or free-field sound sources. Under the present conditions, normal subjects hear two spatially disparate sounds, i.e. one sound image at the left ear and one sound image at the right ear (for further details see [3]). In this second test, 20 monaural sound stimuli, 10 to the left and 10 to the right ear, as well as 10 binaural incoherent sounds were presented following a quasi-random order. On the basis of these two tests, patients were classified as showing extinction when they reported at least 90% of the left or right stimuli on each side correctly, but failed to indicate the left stimulus during bilateral stimulation in more than 50% of the trials. Three patients showed severe auditory extinction in both tests; one patient only when tested under standardized experimental conditions (Table 2). The RBD patients without extinction and the healthy subjects did not show any failures in both tests Monaural hearing test All subjects were tested for unilateral hearing loss. For this purpose, we presented monaural sound stimuli (setting, parameters, and equipment as above). Sound-pressure level was varied following a quasi-random order from 0 to 80 db re 20 Pa, in steps of 10 db. Subjects were instructed to verbally indicate the occurrence of a sound. The procedure consisted of 63 trials and was carried out separately for each ear. Individuals who exhibited differences of more than 10 db in the sensitivities of the two ears were excluded (n = 3). The mean absolute interaural difference in thresholds was 3.2 db (S.D. 0.8) in the four patients with auditory extinction, 1.7 db (S.D. 2.3) in the five RBD patients without extinction, and 3.1 db (S.D. 2.6) in the five healthy controls, with no significant differences between these three groups of subjects (F(2, 13) = 0.742; P = 0.499). In two of the extinction patients, the left ear was slightly more sensitive than the right ear ( 2.8 db, each), while in the other two patients thresholds were lower for the right ear (+4.3 db; +2.7 db). Also no systematic biases in favour of one ear were measured in the two control groups Main procedure The general conditions in the main experiment (setting, equipment, etc.) were as described above. In each trial, sound pulses consisting of incoherent band-pass filtered noise (cutoff frequencies 0.6 and 5 khz; duration 40 ms; rise and fall time 20 ms; 70 db re 20 Pa) were presented via headphones to the left and right ear. The time difference between the envelops of left and right sounds (SOA) was varied between trials following a quasi-random order from 700 ms (left sound leading by 700 ms) to +700 ms (right sound leading by 700 ms). An SOA of zero indicates physical simultaneity of the left and right sounds. Within the range of 400

3 H.-O. Karnath et al. / Neuropsychologia 40 (2002) Table 1 Demographic and clinical data of the groups of brain-damaged patients and of the healthy subjects Auditory extinction RBD patients Healthy subjects Number Sex 3f,1m 2f,3m 2f,3m Age Median (range) 58 (43 76) 55.6 (34 75) 59.8 (51 64) Etiology Infarct 2 4 Hemorrhage 2 1 Lesion Subject number 1 Bg, I, F, T P Subject number 2 I, F P Subject number 3 Bg, I, F, T Bg, I, F Subject number 4 Bg, I P Subject number 5 F, P Time since lesion Days Median (range) 11 (8 13) 9.3 (1 22) Visual field defect Percentage present 0 0 Extinction Visual Percentage present 75 0 Tactile Percentage present 50 0 Neglect Percentage present 75 0 Letter cancellation Left Median (range) 11.5 (0 30) 29.2 (28 30) Right Median (range) 23.8 (16 30) 30 (30) Bells test Left Median (range) 5 (0 14) 14.4 (13 15) Right Median (range) 9.5 (4 15) 15 (15) Baking tray task Left Median (range) 4.5 (1 8) 7.2 (7 8) Right Median (range) 11.5 (8 15) 8.8 (8 9) Copying Percentage Omitted Median (range) 21.9 (0 87.5) 2.5 (0 12.5) RBD: right brain-damaged patients without extinction; sex f: female, m: male; lesion F: frontal, T: temporal, P: parietal, Bg: basal ganglia, I: insula; visual field defect: visual field defects (hemianopia, quadrantanopia) were assessed by standard neurological examination and/or Tübingen perimetry; extinction: for each modality, 10 unilateral stimuli on either side and 10 bilateral stimuli were presented in a quasi-random order. Patients were classified as showing extinction when they reported at least 90% of the unilateral stimuli on each side correctly, but failed to perceive the left stimulus during bilateral stimulation in more than 50% of the trials. Visual extinction was tested by the usual clinical confrontation technique; tactile extinction was investigated by applying short fingertips on the dorsal surface of the patient s left and/or right hand. Spatial neglect: spatial neglect was diagnosed when the patients showed the typical clinical behaviour such as (a), a spontaneous deviation of the head and eyes toward the ipsilesional side, (b) orienting towards the ipsilesional side when addressed from the front or the left, and (c) ignoring of contralesionally located people or objects. In addition, all patients had to fulfill the criterion (cf. [5]) in at least two of the following four tests. Letter cancellation [15]: the number of correct target letters on each half of the test sheet is n max = 30 on either side. Bells test [6]: the number of correct targets on each half of the test sheet is n max = 15 on either side. The five targets of the central column were not regarded. Baking tray task [13]: any distribution that is more skewed than seven items in the left half and nine on the right was considered a sign of neglect. Copying task: the multi-object scene consisted of four elements (a fence, a car, a house, and a tree) two in each half of a DIN A4 sheet of paper: omission of at least one of the left-sided features of each figure was scored as one and omission of one whole figure was scored as two for each omitted figure. One additional point was given when left-sided figures were drawn on the right side. The maximum score was 8. to +400 ms, SOAs were varied in steps of 50 ms, and for longer SOAs in steps of 100 ms. Each session was composed of 206 trials that were presented in two blocks of equal durations, with a rest of 5 min between them. Subjects were Table 2 Percent of sound stimuli not perceived by the patients with auditory extinction when the stimuli were presented unilaterally at the left or right (L/R) ears or bilaterally at both ears Bedside testing Standardized testing Unilateral Bilateral Monaural Binaural L R L R L R L R Patient Patient Patient Patient informed that within each trial two sounds were presented to the left and right ear. They were instructed to make a verbal forced-choice response regarding the sequence of events ( left-then-right or right-then-left ). Trials were presented with intervals of 5 s. In case the subject did not respond within this 5-s interval, the trial was repeated at the end of the session. For each subject, the frequency of the judgements right-then-left was plotted as a function of SOA. The data were fitted to the sigmoidal equation: f(soa) = 100 (1 + e k(soa SOA 50%) ) where f is the frequency of judgements right-then-left, given as percentage; SOA 50% is that SOA where f is 50%; k is the slope of the function at SOA 50% ; e the base of the natural logarithm [8]. That SOA where the proportion of the

4 1980 H.-O. Karnath et al. / Neuropsychologia 40 (2002) subject s judgements right-then-left or left-then-right was 50% (SOA 50% ) characterizes the point of subjective binaural simultaneity, i.e. that time difference at which the left and right sounds were perceived as being presented simultaneous. 3. Results Fig. 1 illustrates the individual psychometric functions of all subjects as well as the mean SOA 50% values obtained for each group. The statistical comparison of the mean SOA 50% revealed a significant difference between the three subject groups (F(2, 13) = 99.98; P < 0.001). Post-hoc testing using a Bonferroni corrected α-level obtained that the mean SOA 50% of the patients with auditory extinction was significantly shifted towards negative values compared with the SOA 50% found in RBD patients without extinction (t (7) = 10.25; P < 0.001) and in the healthy subjects (t (7) = 13.74; P < 0.001). The patients with auditory extinction perceived simultaneity of the two sounds when the sound at the left ear was leading in time, on average, by ms (S.D. 41.4) over the one presented at the right ear. Fig. 1. Perception of SOAs of two stimuli presented to the left and right ear for the groups of brain-damaged patients with auditory extinction, of RBD patients without extinction (RBD), and of healthy subjects. The percentage of judgements right-then-left is plotted as a function of SOA. Solid lines represent the fitted individual psychometric functions from each subject. The coefficients of determination were significant for each subject (R 2 > 0.85, P<0.001), indicating a close fit of each psychometric function to the individual data. Dotted lines indicate the mean SOA per group where judgements right-then-left and left-then-right were 50% (SOA 50% ), i.e. the point of subjective binaural simultaneity. Negative SOAs indicate that the left sound is leading in time; positive SOAs indicate that the right sound is leading.

5 H.-O. Karnath et al. / Neuropsychologia 40 (2002) The SOA 50% values both in RBD patients without extinction (mean 11.0 ms, S.D. 36.3) and in healthy subjects (mean ms, S.D. 21.8) were close to zero. There was no significant difference of SOA 50% values between these two groups (t (8) = 1.44; P = 0.189). 4. Discussion The present study demonstrated that a prior entry bias which has been previously observed for visual and for tactile ipsilesional events in patients with visual and with tactile extinction [1,2,9,11] also characterizes the time-course of awareness for auditory events in patients with auditory extinction. Although previous as well as present findings do not allow to assume a causal relationship between extinction and prior entry, the patients with auditory extinction perceived binaural simultaneity in contrast to both control groups when contralesional sound events were substantially delayed. Thus, a prior entry bias towards the side ipsilateral to the brain lesion appears to be a general phenomenon associated with extinction in the visual, tactile as well as in the auditory modality. It has long been known that sensory events occurring at a location to which spatial attention is deployed will tend to be perceived prior to physically synchronous events at unattended locations (e.g. [14,16]). The observation that patients with extinction perceive ipsilesional events earlier than synchronous contralesional stimuli thus strongly suggests that the brain lesions in these patients lead to a chronic bias of spatial attention towards the ipsilesional side. Observations that favour this assumption have been reported by Karnath [7]. In the first experiment of that study, three patients with visual extinction were asked to name stimuli (geometrical figures such as rectangles, triangles, etc.) presented either unilaterally in the left or the right visual half-field or bilaterally on both sides. The patients were not able to predict the appearance of the stimuli because all three conditions of stimulus presentation were randomized. In the second experiment of that study, only bilateral stimuli were presented and the patients were explicitly informed that on each trial there will be a stimulus on the left as well as on the right side. In both experiments, the patients showed extinction of the contralesional stimuli on bilateral trials. However, in all patients the contralesional stimuli were not entirely extinguished. To a certain percentage, they were able to name both leftand right-sided stimuli in the bilateral trials. Interestingly, in these successful bilateral trials, all patients stereotypically began naming of the ipsilesional figure first (100% of the trials in all patients) and subsequently named the figure presented on the contralesional side. Thus, it was concluded that in patients with extinction there must be a directionally specific, chronic orienting of attention to the ipsilesional side, leading to a primary analysis of ipsilaterally located information [7]. This conclusion is in line with a widely discussed interpretation of the prior entry effect, namely that stimuli occurring at an attended location receive privileged access to awareness because the rate of information processing is enhanced at attended relative to unattended locations (e.g. [12,16]). The present study suggests that the delay of information processing for contralesional events might have a comparable magnitude in different sensory modalities. Our patients with auditory extinction perceived simultaneity of the two sounds when the sound at the left ear was leading in time over the one presented at the right ear by 270 ms. Interestingly, the patients with visual extinction investigated previously [1,11] reported that the ipsilesional visual stimulus preceded the contralesional unless the latter led by over 200 ms. More research is necessary to further investigate whether or not the delay for contralesional events is in fact similar across modalities. But, so far, it seems that the costs for information processing of contralesional events in extinction, induced by the bias of spatial attention towards the ipsilesional side, affect awareness of visual as well as auditory events to a similar degree. Beyond this aspect, the present as well as the previous results reporting a prior entry bias in extinction suggest that awareness is severely delayed for contralesional stimulus events even under conditions where a concurrent ipsilesional competitor is not physically present. When the contralesional sound was presented first, it was treated abnormally with temporal leads over the (subsequently presented) ipsilesional sound as great as 270 ms. This finding accords with the previous observation that identification of contralesional visual stimuli was disturbed in patients with extinction even with exclusive unilateral stimulus presentation, i.e. even when no subsequent stimulus was presented on the ipsilesional side [7]. In the latter study, unilateral visual stimuli were presented either to the left or to the right of a central fixation point. While the identification performance was unimpaired when the stimuli were presented with a conventional duration of 180 ms, the percentage of correct identification responses of the same stimuli dropped substantially on the contralesional side when the difficulty of the task was increased by reducing the duration of stimulus presentation to 10 ms. In line with this finding are more recent studies that tested the effects of temporal asynchrony on extinction. Again, these studies found an impaired processing of contralesional stimuli with temporally isolated events. Di Pellegrino et al. [4] and Baylis et al. [1] reported that their patients with visual extinction were impaired to identify the contralesional stimulus not only with simultaneous presentations but also when it preceded or followed the ipsilesional stimulus. In conclusion, all these findings argue against the interpretation that extinction represents an impairment in disengaging attention from ipsilesional events in order to detect contralesional events [10]. Rather, it seems that the brain lesions in patients with extinction induce a chronic bias of spatial attention towards the ipsilesional side, leading to primary analysis of ipsilaterally located information.

6 1982 H.-O. Karnath et al. / Neuropsychologia 40 (2002) Acknowledgements This work was supported by grants from the Deutsche Forschungsgemeinschaft (Ka 1258/6-1, Ka 1258/2-3, Eh 91/4-4, Gu 261/7-2). We thank P. Dillmann for preparing the software and parts of the electronic equipment. References [1] Baylis GC, Simon SL, Baylis LL, Rorden C. Visual extinction with double simultaneous stimulation: what is simultaneous? Neuropsychologia 2002;40: [2] Birch HG, Belmont I, Karp E. Delayed information processing and extinction following cerebral damage. Brain 1967;90: [3] Blauert J. Spatial Hearing: The psychophysics of human sound localization. Cambridge, MA: MIT Press, [4] Di Pellegrino G, Basso G, Frassinetti F. Spatial extinction on double asynchronous stimulation. Neuropsychologia 1997;35: [5] Ferber S, Karnath H-O. How to assess spatial neglect line bisection or cancellation tasks? Journal of Clinical and Experimental Neuropsychology 2001;23: [6] Gauthier L, Dehaut F, Joanette Y. The Bells test: a quantitative and qualitative test for visual neglect. International Journal of Clinical Neuropsychology 1989;11: [7] Karnath H-O. Deficits of attention in acute and recovered visual hemi-neglect. Neuropsychologia 1988;26: [8] Lewald J, Karnath H-O. Sound lateralization during passive wholebody rotation. European Journal of Neuroscience 2001;13: [9] Mattingley JB. Spatial extinction and its relation to mechanisms of normal attention. In: Karnath H-O, Milner AD, Vallar G, editors. The cognitive and neural bases of spatial neglect. Oxford: Oxford University Press, 2002, in press. [10] Posner MI, Walker JA, Friedrich FJ, Rafal RD. Effects of parietal injury on covert orienting of attention. Journal of Neuroscience 1984;4: [11] Rorden C, Mattingley JB, Karnath H-O, Driver J. Visual extinction and prior entry: impaired perception of temporal order with intact motion perception after unilateral parietal damage. Neuropsychologia 1997;35: [12] Stelmach LB, Herdman CM. Directed attention and perception of temporal order. Journal of Experimental Psychology: Human Perception and Performance 1991;17: [13] Tham K, Tegnér R. The baking tray task: a test of spatial neglect. Neuropsychological Rehabilitation 1996;6: [14] Titchener EB. Lectures on the elementary psychology of feeling and attention. New York: Macmillan, [15] Weintraub S, Mesulam M-M. Mental state assessment of young and elderly adults in behavioral neurology. In: Mesulam M-M, editor. Principles of behavioral neurology. Philadelphia: F.A. Davis Company, p [16] Zackon DH, Casson EJ, Zafar A, Stelmach L, Racette L. The temporal order judgement paradigm: subcortical attentional contribution under exogenous and endogenous cueing conditions. Neuropsychologia 1999;37: