Heightened Awareness of Perceptual Anomaly Relates to Gamma Frequency

1
Heightened Awareness of Perceptual Anomaly
Relates to Gamma Frequency Abnormality in
Schizophrenia J.K. Johannesen1,2, Bismark, A.
W., M.D. Bodkins2, A. Shekhar2, B.F. ODonnell3,
W.P. Hetrick3 1 Psychology, Indiana
University-Purdue University Indianapolis 2
Psychiatry, Indiana University School of
Medicine 3 Psychology, Indiana University
Psychophysiological Analyses
BACKGROUND
METHODS

• P50 Perceptual anomaly SZ participants (M
  1.74, sd .62) produced smaller P50 S1
  amplitudes than perceptually normal SZ (M 2.28,
  sd .30 p lt .05) and healthy normal
  participants (M 2.11, sd .62 p lt .05).
  Perceptually normal SZ did not differ from the
  healthy normal group. All three groups were
  equivalent according to S2 amplitude and
  suppression (S1-S2) values.

Aberrations of perceptual experience represent
a fundamental feature of schizophrenia. Patients
describe heightened awareness, distractibility,
and sensitivity to sensory information (1).
Moreover, individuals with schizophrenia are
thought to experience flooding (2), a condition
in which the brain is inundated by sensory
stimuli without adequate regulation over the
source and quality of sensory inputs. A recent
study reported that approximately 50 of patients
evaluated by structured interview acknowledged
perceptual disturbances thought to reflect poor
gating of sensory stimuli (3). The integrity
of neuronal mechanisms involved in auditory
gating may be indexed by the cortical
event-related potential (ERP) evoked by identical
paired click stimuli. Under this paradigm, the
P50 ERP to the second stimulus (S2) is normally
reduced in amplitude relative to the first (S1).
Poor attenuation of the P50 response is widely
observed in schizophrenia (4), and interpreted to
suggest weak neuronal gating of the redundant
stimulus (i.e., S2). However, efforts to relate
individual differences in perceptual
phenomenology to psychophysiological measurement
(P50) have yet to provide a clear link between
these theoretically integrated domains of
impairment in schizophrenia (5). The P50 ERP
is comprised of activity from two distinct
frequency bands (6) a gamma-band response (GBR
20 to 50 Hz) and a low-frequency response (LFR 1
to 20 Hz). While the GBR is associated with the
initial neural registration of sensory stimulus
(7), the LFR reflects early stages of attention
processing (8) and information encoding (9).
Therefore, analysis of the P50 ERP by these
component frequency domains may enable closer
examination of specific functional relationships.
The present study examined phenomenological and
physiological (P50, GBR, LFR) indices of sensory
processing to address three principle hypotheses
regarding their theoretical relationships in
schizophrenia.
Sample Twenty-six DSM-IV (American Psychiatric
Association, 1994) diagnosed SZ 42.2 years old
(SD 9.2), 65 male and 32 healthy normal 42.4
years old (SD 8.8), 47 male participants were
evaluated. All SZ participants were prescribed
atypical antipsychotic medication (average
Chlorpromazine EQ dose 550 MG daily, SD
462.7). Clinical measures Referral diagnosis
was confirmed using the Structured Clinical
Interview for DSM-IV Axis I Disorders(SCID-IP).
Subjective appraisal of sensory disturbance was
assessed in all participants using the Sensory
Gating Inventory (SGI10). Respondents rated 36
items using a 6-point Likert-type scale, ranging
from 0 (never true) to 5 (always
true). Sample SGI items My hearing is so
sensitive that ordinary sounds become
uncomfortable At times I have trouble focusing
because I am easily distracted I notice
background noises more than other people.
Electrophysiological measure EEG was acquired
through a 32-channel cap and amplification system
at a rate of 1K Hz. One-hundred ten paired clicks
(500 ms ISI 7-11 s ITI) were delivered
binaurally with a peak click intensity of 81 dB
SPL and click duration of 3 ms against a 58 dB
SPL white-noise background. Data was subjected to
an ICA eye-blink correction and spatial principle
component analyses. A fronto-central spatial
component was extracted for ERP analysis using
two data processing methods (1) conventional
peak picking of the P50 ERP in data low-pass
filtered at 50 Hz, and (2) spectral power (FFT)
analysis within the low- (1-20 Hz) and gamma-band
(20-50 Hz) frequency ranges. Statistical
Analysis ERP values were square root
transformed. One-way ANOVA was used to compare
groups on ERP amplitudes and suppression,
calculated using the S1-S2 difference score. SGI
mean scores were entered into logistic regression
to predict diagnostic class using a
classification cut-off of .50. ERP comparisons
were reassessed according to predicted group
classifications using LSD post-hoc tests.

• LFR No differences were observed according to
  LFR magnitude or suppression values

RESULTS

• SZ participants, as a group, endorsed
  significantly more perceptual disturbance than HC
  (Table 1).
• Comparison across ERP values by diagnostic class
  yielded significance only with respect to the
  LFR, indicating lower magnitude S1 responses in
  SZ (Table 1).
• The SGI, entered into logistic regression,
  accurately classified 79.3 of the study sample
  cumulatively (Wald 12.652, beta 1.49, df 1,
  p lt .001). Specificity (accurate classification
  of HC) was 90.6 and sensitivity (accurate
  classification of SZ) was 65.4 (Table 2).

• GBR Perceptual anomaly SZ produced smaller
  magnitude GBR S1 magnitude responses (M .13, sd
  .07) than perceptually normal SZ (M .20, sd
  .06 p lt .05). Perceptual anomaly SZ exhibited
  less GBR suppression (M .03, sd .05) than
  perceptually normal SZ (M .08, sd .04 p lt
  .05) and healthy normal participants (M .06, sd
  .05 p lt .05). Perceptually normal SZ were
  comparable to healthy participants in each
  comparison.

Diagnostic Group Comparisons on SGI and ERP Values
Classification Based on SGI
CONCLUSIONS

• References
• McGhie A, Chapman, J (1961) British Journal of
  Medical Psychology 34103-116.
• Venables P H (1964) Progress in Experimental
  Personality Research 721-47.
• Bunney WE., Hetrick W P, Bunney B. et al.
  (1999) Schizophrenia Bulletin 25 577-592.
• Adler LE, Pachtman E, Franks R D, et al. (1982)
  Biological Psychiatry 17639-654.
• Jin Y, Bunney WE, Sandman CA, et al (1998)
  Biological Psychiatry 43873-8.
• Clementz BA, Blumenfeld LD (2001) Experimental
  Brain Research 139377-90.
• Karakas S, Basar E (1998) International Journal
  of Psychophysiology 3113-31.
• Naatanen R, Teder W (1991) Acta Oto
  Laryngologica Supplement 491161-6.
• Klimesch W (1999) Brain Research Brain Research
  Reviews 29169-95.
• Hetrick WP, Smith D (Manuscript under review).

• The Sensory Gating Inventory provides a highly
  specific method of discriminating SZ from healthy
  normal populations, however, perceptual anomaly
  may not be universally or consistently
  experienced in SZ.
• Poor LFR activation is a prominent feature SZ but
  does not appear to relate to the severity of
  perceptual disturbances
• The GBR, previously found to be comparable
  between SZ and HC (6), was uniquely impaired in a
  subgroup reporting high levels of perceptual
  disturbance and appears to contribute to smaller
  P50 S1 amplitude in this group
• Inconsistent with a recurrent inhibitory gating
  model, observed deficits seem less the result of
  poor gating of S2 than weak activation of a
  response to S1

Hypotheses

• Based on previous reports (6), disturbance in low
  frequency activity will best distinguish
  schizophrenia (SZ) from healthy control
  participants
• Only a portion of the SZ sample (est. 50) will
  be deviant with respect to the normal range of
  perceptual experience
• Based on theoretical rationale, disturbance in
  gamma band activity will best distinguish SZ
  participants reporting aberrant vs. normal
  perceptual experience

• SZ participants classified as SZ (SZ-SZ
  hereafter perceptual anomaly SZ) provided
  higher SGI ratings (M 3.17, sd .74) than SZ
  participants classified as HC (SZ-HC M 1.21,
  sd .54 p lt .001) and healthy normal
  participants (M 1.17, sd .47 p lt .001). The
  latter two groups provided nearly equivalent SGI
  ratings (p .847), thus, the SZ-HC group is
  hereafter referred to as perceptually normal SZ
  .