Neuropsychology of PTSD

Darren Weber, BSc(Hons), BA

Neuropsychological Theory

Kolb (1987) proposed a neuropsychological explanation of intrusive memories and associated psychophysiological arousal in PTSD.  He has supplemented a two-factor learning theory that PTSD results from both classical conditioning of extreme emotional responses to traumatic stimuli (i.e., fear, terror, anger, rage, sadness, guilt, and indignation) and operant conditioning of emotional numbing, withdrawal, and avoidance of traumatic stimuli (Kolb, 1987).  He proposed that PTSD results from excessive traumatic stimulation that overwhelms the capacity to process information efficiently (Kolb, 1987).

Such stimulus overload occurs when the ... capacity to process information signalling threat to life overwhelms the cortical ... processes concerned with perceptual discrimination and effective adaptive responses for survival (Kolb, 1987, p. 993).

Excessive threatening stimulation sensitises neuronal circuits that are responsive to threat, including structures in the “temporal-amygdaloid complex concerned with agonistic behaviour”, which may “lead to depression [or atrophy] of those synaptic processes which permit habituation and thus discriminative perception and learning” (Kolb, 1987, p. 993).  As a result of this initial threatening sensitisation and a diminished capacity for discrimination and habituation, sensitivity to threatening stimuli generalises, via a process of higher order conditioning, from an initial traumatic experience to a variety of similar threatening situations (Kolb, 1987).  Furthermore, Kolb (1987) proposes that this generalised sensitivity to threatening situations disrupts cortical inhibition of the locus ceruleus and its activation of diverse cortical and subcortical structures.  The consequent increase in cortical and subcortical activation further enhances the sensitisation and activation of the perceptual, cognitive, affective, and somatic processes associated with threatening situations (Kolb, 1987).  As a result,

in the face of perceived threats there occurs excessive sympathetic arousal - including neuroendocrine disturbances as well as behavioural expressions of rage and irritability and repetitive cortical reactivation of memories related to the traumatic events.  The latter are projected in the daytime as intrusive thoughts and at nighttime in the recurrent traumatic nightmares of PTSD ... (Kolb, 1987, p. 994).

Thus, Kolb (1987) proposed that PTSD reflects a vicious cycle of threat related stress and hyperactivity.  Firstly, a traumatic experience overwhelms the cognitive system and promotes sensitivity to threatening stimulation that disrupts the capacity to process information efficiently.  Secondly, this sensitivity to threat activates a stress response that consists of a vicious cycle of central and peripheral adrenergic hyperactivity.  Thirdly, feedback from this stress response enhances the cognitive, affective, somatic, and behavioural responses to threat, which produces both reactivation of threat related feelings and memories and chronic autonomic hyperarousal (see also Burges-Watson, Hoffman, and Wilson, 1988).

Similarly, Everly (1989, 1993) proposed that PTSD comprises a psychological sensitivity to threat and a concomitant hyperactive stress response.  Intense threatening stimulation can promote sensitivity to threatening information and a concomitant hyperactivity of stress related responses (Weil, 1974; cite in Everly, 1993, p. 273).  Although a sensitivity to interpret ambiguous stimuli as threatening and to activate intense and rapid responses to stress are adaptive in dangerous situations, they are inappropriate and debilitating in otherwise normal situations.  Everly (1993) proposes that the cognitive sensitivity to threat is primarily instantiated in frontal and cingulate cortical areas, which activate an affective and somatic stress response instantiated in limbic structures, the sympathetic nervous system, and the hypothalamic-pituitary-adrenal structures of the neuroendocrine system.  Furthermore, activation of the limbic system and components of the somatic stress response has a positive feedback effect on the cognitive and affective sensitivity to threat.

. . . affective discharge from the limbic brain sends neural impulses in two simultaneous directions: 1) to neocortical targets, and 2) to the skeletal musculature . . . .  The neocortical centers then send impulses back to the limbic areas, thus sustaining affective arousal.  At the same time, proprioceptive impulses from the skeletal musculature ascend via the reticular formation and further stimulate limbic and neocortical targets.  This complex positive feedback loop serves to sustain and intensify [threat related arousal] (Gellhorn, 1965, 1967; cited in Everly, 1993, p. 273).

Furthermore, this hyperarousal has long-term neurological consequences; intense or protracted stimulation produces various structural and functional changes in networks of the limbic system, including neuronal atrophy in the hippocampus and amygdala (Weil, 1974; Post, 1985, 1986, 1992; cited in Everly, 1993, p. 273-274; see also Bremner et al, 1995).  Everly (1989) proposed that tonic or phasic hyperactivity of the noradrenergic system in the septo-hippocampal-amygdaloid formation explains many facets of the phenomenology of PTSD, including heightened startle responses and autonomic hyperactivity, emotional lability, irritability, fear, guilt, aggression, and intrusions such as flashbacks and nightmares.  The noradrenergic system of the septo-hippocampal complex is involved in both the integration of novel or aversive stimulus information and the concomitant activation of amygdaloid and hypothalamic-pituitary responses to threat (Gray, 1982; cited in Everly, 1989, p. 316).  Also, activity in the septo-hippocampal-amygdaloid network is involved in memory and panic or fear reactions (Reiman et al, 1986; Gloor, 1986; Post, 1986, Seifert, 1983; cited in Everly, 1989, p. 316).  Moreover, noradrenergic activation of the septo-hippocampal-amygdaloid system both inhibits its accommodation or habituation (Madison and Nicoll, 1982; cited in Everly, 1989, p. 316) and sensitises it and facilitates its response to further novel or aversive stimuli (Gray, 1982; cited in Everly, 1989, p. 316).  Thus, Everly (1989) hypothesized that PTSD results from noradrenergic hyperactivity in the septo-hippocampal-amygdaloid system (see also Burges-Watson, Hoffman, and Wilson, 1988).

Neuropsychological Assessment

Everly and Horton (1989) hypothesised that adrenergic hyperactivity in PTSD promotes hypersensitivity of neural circuits in the limbic system, which disrupts memory processes instantiated in septo-hippocampal structures.  To test this hypothesis, they presented 14 PTSD patients with the Four-Word Short-Term Memory Test (Ryan and Butters, 1980; cited in Everly & Horton, 1989, p. 808).  This test requires subjects to: (a) listen to four unrelated words, (b) count backwards in multiples of three from a random three digit number for 15 or 30 seconds, and (c) recall the previous four words.  After 15 seconds of counting, PTSD patients had a mean recall rate of only 35% of the words presented.  After 30 seconds of counting, PTSD patients had a mean recall rate of only 19% of the words presented.  These results provide tentative confirmation of their hypothesis that there is short-term memory impairment in PTSD.  However, their study was flawed by a lack of an appropriate control group and the concomitant application of appropriate statistical analyses.

On the basis of clinical reports and the tentative results of Everly and Horton (1989), Gil, Calev, Greenberg, Kugelmass, and Lerer (1990) hypothesised that there are attention and memory deficits in PTSD patients.  They evaluated the performance of 12 PTSD patients, 12 psychiatric controls, and 12 normal control subjects on several neuropsychological tasks.  PTSD patients were male combat veterans and victims of terrorism or vehicle accidents who were drug-free for at least two weeks and no subjects abused alcohol or other substances.  Before a traumatic experience, PTSD patients had a similar IQ to that of controls, but their IQ deteriorated quicker than controls after a traumatic experience.  As often noted in clinical reports, PTSD patients reported subjective awareness of more difficulties with attention and memory than controls.  In fact, their attention was impaired - they detected fewer targets than controls in a continuous performance task that required the detection of the products of the number seven in a tape recording of a series of numbers.  Furthermore, PTSD patients exhibited deficits in frontal, executive functions - they had poor letter and category verbal fluency, poor performance of the comprehension, similarities, and digit-symbol subscales of the Wechsler Adult Intelligence Scale (WAIS, Wechsler, 1955; cited in Gil et al., 1990, p. 32), and poor performance of automatic tasks of the mental control subscale of the Wechsler Memory Scale (WMS, Wechsler and Stone, 1945; cited in Gil et al., 1990, p. 32; e.g., counting backwards, counting forwards in multiples of three, and reciting the alphabet).  Also, PTSD patients had deficits of verbal and visual immediate memory - they had poor recall for (a) the digit span subtest of the WAIS, (b) the difficult paired-associated task from the WMS, and (c) the Bender-Gestalt Test (Hutt, 1977; cited in Gil et al., 1990, p. 32) and the Benton Visual Reproduction Test (Gil et al., 1990, p. 33).  Furthermore, PTSD patients had poor long-term memories for episodes of the Famous Events Questionnaire (Calev et al 1995; Squire et al, 1979; cited in Gil et al., 1990, p. 34).  There were no differences in performance between PTSD patients and other psychiatric controls.  These results of Gil et al. (1990) clearly indicate deficits in attention and memory processes in PTSD patients, but the impairments cannot be attributed to a traumatic experience alone.

Uddo, Vasterling, Brailey, and Sutker (1993) examined attention, memory, and learning in 16 Vietnam veterans with chronic PTSD and 15 military personnel, matched for sex, race, and education.  Military personnel were slightly younger than PTSD patients, but analyses of covariance revealed no significant influence of age on their results.  Their subjects were excluded on the basis of prior psychosis, substance abuse, neurological pathology, head injury or loss of consciousness, and scores below 85 on the WAIS-R or Shipley Institute of Living Scale (Shipley, 1967; cited in Uddo et al., 1993, p. 45).  They found several indications of attention and immediate memory impairments in PTSD patients.  PTSD patients had poor category verbal fluency and they had poor acquisition, perseveration, and sensitivity to proactive interference, but no influence of retroactive interference and no deficit in delayed memory in the Auditory Verbal Learning Test (AVLT, Rey 1964; cited in Uddo et al., 1993).  PTSD patients had poor immediate memory, but normal delayed memory for the Complex Figure Test (CFT, Osterreith, 1944; cited in Uddo et al., 1993) and poor performance of the visual memory span subtest of the WMS-R (Wechsler, 1985; cited in Uddo et al., 1993).  Uddo et al. (1993) suggest that this pattern of cognitive deficits is indicative of frontal-subcortical dysfunction.

Bremner et al (1993) provided further evidence of disturbed information processing in PTSD.  They evaluated the performance of 26 male combat veterans with chronic PTSD against 15 matched control subjects for the logical and figural memory tasks of the WMS and the verbal and visual components of the Selective Reminding Task (SRT, Hannay and Levin, 1985; cited in Bremner et al, 1993).  They controlled for a variety of variables that affect memory performance, various neurological conditions, and the medication status of PTSD patients.  PTSD patients had normal WAIS-R IQ’s.  However, the immediate and delayed recall of PTSD patients was impaired for the verbal and visual material of the SRT and the logical memory task of the WMS.  Their immediate and delayed recall for the figural memory task of the WMS was below normal. but not quite significantly so, although their percentage of retention for that task was negatively correlated with the severity of PTSD symptoms.  Bremner et al (1993) conclude that the verbal and visual, immediate and delayed memory impairments observed in PTSD patients may be attributed to poor concentration and pathology of the hippocampal formation.  They argue that low volume of the hippocampus in PTSD patients may impair its role in memory processes (see also Bremner et al, 1995).

Yehuda et al. (1995) report a further evaluation of auditory verbal learning in PTSD.  They assessed performance of the California Verbal Learning Test (CVLT, Delis, Kramer, Kaplan, Ober, 1987; cited in Yehuda et al., 1995) in 20 male combat veterans with PTSD and 12 normal subjects, matched for sex, race, age, education, and WAIS-R IQ’s.  Patients were excluded on the basis of concurrent psychiatric diagnosis, psychotropic medication and substance abuse, and neurological disorder, brain injury, and loss of consciousness.  PTSD patients had normal acquisition and cumulative learning for an initial list of words in the CVLT and patients were not affected by proactive interference, but they were sensitive to retroactive interference in the CVLT.  Yehuda et al. (1995, p. 139) propose that this memory impairment reflects a deficit in the executive monitoring or regulation of episodic memory, such that “memory dysfunction in PTSD may involve the intermingling of past experiences with current experiences (e.g., in the form of intrusive thoughts or flashbacks).”

Evaluation and Conclusions

Neuropsychological theory proposes that hyperactivity of the noradrenergic networks of the septo-hippocampal formation is associated with sensitivity to threat and a susceptibility to recurrent traumatic intrusions, which impairs perceptual discrimination and memory processes.  Systematic investigation of neuropsychological performance in PTSD provides support for this theory; it has identified deficits of attention and memory, although some results are equivocal (see Table 1).

PTSD patients have deficits of attention and frontal, executive functions (Gil et al., 1993; Uddo et al., 1993).  These deficits are apparent in the performance of various tasks, including a continuous performance task, verbal fluency, comprehension, similarity judgements, digit-symbol substitution, and automatic tasks such as counting backwards, counting forwards by threes, and reciting the alphabet.

PTSD patients also have memory impairments.  Several studies have identified immediate memory impairments in verbal and visual tasks (Bremner et al, 1993; Everly & Horton, 1989; Gil et al., 1993; Uddo et al., 1993).  Some studies have identified long-term memory deficits for verbal tasks and famous events (Bremner et al, 1993; Gil et al., 1993), but results for visual tasks are equivocal (Bremner et al, 1993; Uddo et al., 1993).

Two studies that have evaluated auditory verbal learning in PTSD provide equivocal results. Uddo et al. (1993) have found deficits in acquisition, sensitivity to proactive interference, and no influence of retroactive interference in PTSD.  However, Yehuda et al. (1995) have found normal acquisition, no influence of proactive interference, but sensitivity to retroactive interference in PTSD.  The demographic characteristics of the subjects in these studies are similar and both studies controlled for the effects of sex, race, age, education, and WAIS-R IQ’s on performance.  Although the CVLT varies slightly from the AVLT, an important difference between these studies lies in their operational definitions of acquisition and proactive and retroactive interference.  Uddo et al. (1993) defined acquisition as the total words recalled from the first list, whereas Yehuda et al. (1995) defined acquisition as the difference in recall of words between the first and fifth presentation of the first list. The acquisition results of Uddo et al. (1993) have confounded an initial attention and immediate memory component (evident in recall after the first presentation) and a learning component (the degree of improvement in recall over subsequent presentations).  The effect of this operational difference would be to exaggerate the recall deficit in PTSD found by Uddo et al. (1993).  However, the finding of Uddo et al. (1993) has a robust significance level, suggesting a deficit in PTSD for at least one of these components.  To evaluate proactive and retroactive interference, Uddo et al. (1993) used a ratio of the number of words recalled from both lists, whereas Yehuda et al. (1995) subtracted the number of words recalled in the second list from the first list.  However, this operational difference may only produce minor differences in the interference results of the two studies.

Table 1.  A Summary of Neuropsychological Evaluations of PTSD

Note: California Verbal Learning Test (CVLT), Rey Auditory Verbal Learning Test (AVLT), Rey Complex Figure Test (CFT), Selective Reminding Task (SRT), Wechsler Memory Scale (WMS).

References

Bremner, J., Randall, R., Scott, T., Bronen, R., Seibyl, J., Southwick, S., Delaney, R., McCarthy, G., Charney, D., Innis, R. (1995).  MRI-based measurement of hippocampal volume in patients with combat-related posttraumatic stress disorder. American Journal of Psychiatry, 152: 973‑981.

Bremner, J., Scott, T., Delaney, R., Southwick, S., Mason, J., Johnson, D., Innis, R., McCarthy, G., Charney, D. (1993).  Deficits in short-term memory in posttraumatic stress disorder.  American Journal of Psychiatry, 150: 1015-1019.

Burges-Watson, I., Hoffman, L., Wilson, G. (1988).  The neuropsychiatry of post-traumatic stress disorder.  British Journal of Psychiatry, 152: 164-173.

Everly, G. (1989).  A Clinical Guide to the Treatment of the Human Stress Response.  Plenum Press: New York, pp. 311-322.

Everly, G. (1993).  Psychotraumatology: A two-factor formulation of posttraumatic stress.  Integrative Physiological and Behavioural Science, 28: 270-278.

Everly, G., Horton, A. (1989).  Neuropsychology of posttraumatic stress disorder: a pilot study.  Perceptual and Motor Skills, 68: 807-810.

Gil, T., Calev, A., Greenberg, D., Kugelmass, S., Lerer, B. (1990).  Cognitive functioning in post-traumatic stress disorder.  Journal of Traumatic Stress, 3: 29-45.

Kolb, L. (1987). A neuropsychological hypothesis explaining posttraumatic stress disorder.  American Journal of Psychiatry, 144: 989-995.

Uddo, M., Vasterling, J., Brailey, K., Sutker, P. (1993).  Memory and attention in combat-related post-traumatic stress disorder.  Journal of Psychopathology and Behavioural Assessment, 15: 43-52.

Yehuda, R., Keefe, R., Harvey, P., Levengood, R., Gerber, D., Geni, J., Siever, L. (1995).  Learning and memory in combat veterans with posttraumatic stress disorder.  American Journal of Psychiatry, 152: 137-139.