The symptoms of PTSD are indicative of a disturban ce of the normal capacity to resolve cognitive and emotional responses to traumatic events (Yehuda and McFarlane, 1995). Theoretical accounts of PTSD explain the intrusions, avoidance, and heightened arousal and startle responses in terms of the sensitivity of cognitive, affective, and somatic stress response systems to threat. Cognitive, neuropsychological, and psychophysiological assessments of PTSD have revealed information processing biases for threatening information and concomitant abnormalities of attention and memory for non-threatening information.
PTSD is essentially a disorder that involves intrusive recollections, flashbacks, and nightmares related to an extremely threatening experience and endeavours to avoid the disturbing intrusions and associated stress responses. For many people, some emotional processing leads to resolution of emotional reactions to a stressful experience and control over the amount of subsequent attention and thought about the experience. However, for PTSD sufferers, their intense cognitive and affective responses to the traumatic experience are overwhelming, provoking dissociation and avoidance of the distress and great difficulty in resolving their trauma. Nevertheless, stimuli associated with the trauma and a drive to realise some meaning and purpose in experience prompt recollections and dreams about the trauma, but the intense distress of such recollections provokes further dissociation and avoidance. This cycle of recollection and avoidance leaves cognitive and affective responses to the traumatic experience unresolved, so they continue to interfere with their lives for a long time after the traumatic event (Horowitz, 1976; cited in Thrasher, Dalgleish, and Yule, 1994).
The intensity of an exposure to a traumatic, highly threatening event that is uncontrollable or unavoidable could create a highly acute episodic memory of salient features of the trauma (Foa, Steketee, and Olasov-Rothbaum, 1989). Furthermore, the persistent intrusions of the traumatic images, feelings, or reactions are part of the development and elaboration of a sensitive, complex, threat-related information processing schema or "fear network" (Chemtob, Roitblat, Hamada, Carlson, and Twentyman, 1988; Lang, 1977, 1979, 1985, cited in Litz and Keane, 1989). The threat schema could include fragments of sensory, affective, semantic, intentional, or motor information about the initial experience and the circumstances of intrusions. This threat schema may be elaborate, coherent, and stable in PTSD and any events or situations that activate it promote a fear response and further elaborate its content and reinforce its importance to the goal direction and initiative of a PTSD sufferer (Chemtob et al, 1988). Activation of a threat schema encourages PTSD patients to interpret their experience in terms of that schema and also encourages their expectancy that threatening events will occur (Chemtob et al, 1988). This means that PTSD patients are susceptible to interpreting ambiguous events as threatening, which could disturb the continuity of their awareness and distract their attention from their current activities toward extraneous, irrelevant stimuli (Chemtob et al, 1988; Litz and Keane, 1989). This susceptibility to threat interrupts the appropriate development of more adaptive information processing structures for dealing with normal, non-threatening circumstances;
. . . threat arousal inhibits the operation of other information-processing modes or schemata, thereby preventing their operation and further narrowing the attentional focus on threat-related stimuli (Chemtob et al, 1988, p. 266).
This contemporary cognitive theory echoes the thought of Pierre Janet, a pioneer of traumatic research, who described the traumatic recollections of PTSD as an attachment to the trauma that cannot be resolved and replaced by new experiences; he proposed that because PTSD sufferers are
unable to integrate traumatic memories, they seem to have lost their capacity to assimilate new experiences as well. It is . . . as if their personality which definitely stopped at a certain point, cannot enlarge any more by the addition or assimilation of new elements. (Janet, 1911, p. 532, cited in van der Kolk, Herron, and Hostetler, 1994, p. 584).
Janet proposed that cognitions about the traumatic experience cannot be integrated into a complex array of normal experiences; they remain encapsulated within their own cognitive arena, dissociated from a plethora of memories, feelings, and thoughts about normal experiences (van der Kolk, Herron, and Hostetler, 1994). When traumatic cognitions are activated they captivate cognitive resources and deplete the ability to focus on present experiences (van der Kolk, Herron, and Hostetler, 1994). Thus, when cognitions about the trauma are activated they overwhelm the cognitive system, awareness related to them cannot be integrated with the whole personality, and the PTSD sufferer experiences both intrusive distress and irritability and a loss of a sense of personal wholeness and continuity (van der Kolk, Herron, and Hostetler, 1994).
There has been increasing research interest in the cognitive sensitivity to threat in PTSD and a moderate literature on the subject has accumulated. Cognitive assessments of a sensitivity to threatening information have evaluated attention and memory for threatening information and information with neutral and positive or negative emotional valence. Below is an account of that literature and a summary and evaluation of its conclusions and theoretical significance.
Trandel and McNally (1987) investigated unconscious threat-related information processing in 15 Vietnam combat veterans with PTSD, 15 Vietnam veterans without combat experience but an alcohol dependence, and 15 Vietnam veterans without combat experience; all groups were matched for sex, race, age, and education. Subjects were required to shadow words presented to the right ear in a dichotic listening task consisting of a simultaneous series of unrelated words. The task consisted of unrelated words rather than a sentence or story to avoid semantic associations that would enhance unconscious processing and to restrict the ability of subjects to shift attention to the left ear without an interruption to shadowing in the right ear. They presented a series of neutral words to the right ear and the words presented to the left ear consisted of neutral words and four groups of seven critical words: threatening words particularly alarming to Vietnam veterans, words phonemically similar to those threat words, threatening words for agoraphobics and obsessive-compulsives, and neutral words. Trandel and McNally (1987) proposed that if subjects correctly shadow a word in the right ear while presented with a threatening word in the left ear, but commit an error in shadowing the next word in the right ear, there is no shift in conscious attention to the threatening word, but a degree of unconscious semantic processing that interferes with shadowing the following word. Their results indicated that PTSD patients had only a few more shadowing errors than other subjects after the presentation of threatening words, but stringent analyses revealed no significant group differences in shadowing errors. Trandel and McNally (1987) also hypothesised that automatic, unconscious semantic processing of threatening words should be associated with autonomic arousal, but there were no significant group differences in skin conductance after the presentation of threatening words that were not associated with concurrent shadowing errors. These findings may be partially attributed to the low sensitivity of their study; they presented only seven threatening words among a series of 3, 600 words (see also Litz and Keane, 1989).
It is difficult to assess automatic semantic processing in dichotic listening tasks (Trandel and McNally, 1987). Trandel and McNally (1987) assessed unconscious interference with shadowing, but the requirement that unconscious processing should not be associated with errors in shadowing the concurrent word presentation strictly limits the evaluation of automatic processes. It may be possible that automatic semantic processing interferes with concurrent shadowing by provoking conscious awareness of meaning, so a strict constraint of the evaluation of unconscious automatic semantic processing is required in the dichotic listening task. Although PTSD patients may not be more susceptible to unconscious processing of threatening material, this does not imply that they are not susceptible to automatic processing of threatening information. Sensitive threat-related sensory, perceptual, or semantic representations will respond automatically to threatening information and facilitate conscious awareness of that information. In threatening situations, such as combat, automaticity of complex reflex actions and diversion of cognitive resources to deal with danger can make the difference between life or death.
Another method for the assessment of automatic processing of threatening information is a modified Stroop task that involves the presentation of words with various affective associations rather than the names of colours. Automatic semantic analysis of the words provokes semantic associations and consequent cognitive or emotional processing, which interferes with the primary task of naming the colour of the words. Stroop interference related to threatening information reflects automatic activation of threat-related semantic or affective associations and thereby both awareness of that information and avoidance processes that require cognitive resources that are otherwise available for the primary task.
McNally, Kaspi, Riemann, and Zeitlin (1990) evaluated affective Stroop interference in Vietnam combat veterans with (n = 15) and without (n = 15) PTSD. Although veterans with PTSD were younger, less educated, and had more combat exposure than veterans without PTSD, none of these variables were significantly associated with any dependent measures. Veterans were required to name the colours of five different word types on separate sheets. The first sheet consisted of a series of non-words (i.e., "OOOOO"), the second sheet consisted of neutral words (i.e., mix, millionaire, fingertips, concrete, and input), the third sheet consisted of positive words (i.e., love, pleasant, loyal, happy, and friendship), the fourth sheet consisted of words that provoke obsessive-compulsive patients (i.e., germs, filthy, faeces, urine, and dirty), and the last sheet consisted of threatening words associated with the Vietnam war (i.e., bodybags, ‘nam, firefight, medevac, and charlie). The first sheet was used to control for individual variability in colour naming, independently of any semantic interference. Veterans with PTSD were slower to respond to war specific words than words related to obsessive-compulsive disorder and words of neutral or positive affect, whereas veterans without PTSD responded equally to all words. Also, veterans with PTSD were slower to respond to war specific words than veterans without PTSD. Furthermore, the response latency for war specific words was significantly associated with the severity of symptoms in veterans with PTSD and several emotional attributes (i.e., tension, depression, anger, fatigue, confusion, and vigour), but there was no significant association between the ratings of the emotional impact of the words and response latencies for each word type. Thus, PTSD symptoms and concomitant emotional disturbance are associated with a sensitivity to semantic processing of the threat content of information related to the Vietnam war.
Foa, Feske, Murdoch, Kozak, and McCarthy (1991) examined affective Stroop interference in rape victims with chronic PTSD. They assessed rape victims with (n = 15) and without (n = 13) PTSD and normal controls (n = 16); all subjects were female and they were matched for age, WAIS-R vocabulary, several variables associated with the trauma (i.e., injury, use of a weapon, and perceived threat), and the time since trauma (M = 9 months). Subjects were presented with a series of single stimuli in five colours (i.e., white, orange, red, green, and blue) on a computer screen, consisting of non-words (i.e., gosp, narvos, shet, rupe, punic, chorry, peuch, scroam, mulon, and gailt) and words that were rape-related (i.e., rape, assault, stalker, scream, struggle, trapped, V.D., penetrate, nightmare, and attack), generally threatening (i.e., anxiety, death, cancer, tumor, stress, funeral, panic, coffin, guilt, and nervous), or semantically related neutral (i.e., banana, cherry, grape, raisin, apple, prune, peach, strawberry, melon, and pear). Words were matched for length and perceived frequency. Rape victims with PTSD were slower to respond to rape-related words than other words, whereas other subjects responded equally to all words. Also, rape victims with PTSD were slower to respond to rape-related words than both rape victims without PTSD and normal subjects. Thus, rape victims with PTSD have a sensitivity to semantic processing of information related to their trauma.
Cassiday, McNally, and Zeitlin (1992) also examined affective Stroop interference in rape victims with chronic PTSD. They assessed victims of sexual assault with (11 women, 1 man) and without (12 women) PTSD, and a non-traumatised control group (11 women, 1 man); all subjects were matched for race, sex, age, and education and the mean time since trauma for both trauma groups was nine years and five months (range of 4 months to 34 years). Subjects were presented with words of similar length in five colours (i.e., white, yellow, blue, pink, and green) from four word types: high threat (i.e., rape, penis, AIDS, victim, and intercourse), moderate threat (i.e., emergency, crime, lock, flaccid, and bruises), positive (i.e., love, friendship, pleasant, loyal, and happy), and neutral (i.e., polite, moderate, clever, typical, and fair). These words were randomly presented in either a mixed manner or blocked according to word type (i.e., neutral, positive, moderate-threat, and high-threat, in that order). Responses to the random series were slower than a block of stimuli for neutral and positive words, but responses to high and moderate threat words were unaffected by the presentation format. Rape victims with PTSD responded slower to high-threat words than moderate-threat and positive words, which they were slower to respond to than neutral words. Also, rape victims with PTSD responded slower to high-threat, moderate-threat, and positive words than all other subjects. Similarly, rape victims without PTSD were slower to respond to high-threat words than moderate-threat, positive, and neutral words and they were slower to respond to highly threatening words than non-traumatised subjects. Furthermore, response latency for high-threat words in rape victims with and without PTSD was significantly associated with the intrusion subscale, but not the avoidance subscale of the Impact of Events Scale. Many rape victims without PTSD had previously qualified for a diagnosis of PTSD and the results indicate that they have a residual sensitivity for stimuli related to their traumatic experience, suggesting that the cognitive schemata associated with a traumatic experience can remain coherent for some time after the remission of PTSD symptoms. However, since the information processing bias for threatening information is associated with the degree of affective or cognitive intrusion, the primary symptom of PTSD, it is not surprising that the sensitivity or coherence of the threat-related representations is greater in rape victims with chronic PTSD than rape victims without PTSD.
Bryant and Harvey (1995) assessed affective Stroop interference in victims of motor vehicle accidents (MVAs) with PTSD or simple phobia. They evaluated 15 PTSD patients, 15 simple phobia patients, and 15 victims of MVAs without any concomitant morbidity; all subjects were matched for sex, age, WAIS-R vocabulary, severity of MVA, and time since MVA. Subjects were presented with a random series of MVA related words rated as highly and mildly threatening and words of neutral and positive affect on a computer monitor while their time to name the colour of the words was recorded. PTSD patients were slower to respond to highly threatening words than mildly threatening words and words of both neutral and positive affect, while no differences in response latency among the different word types were found in simple phobic or normal subjects. Also, PTSD patients were slower to respond to highly threatening words than both simple phobics and normal subjects. Thus, MVA victims with PTSD exhibited a sensitivity to semantic interference of colour naming for highly threatening information, but not for mildly threatening information. So, although these results support the hypothesis that PTSD patients have a highly aroused threat-related complex, which is different from that of other anxiety disorders, the generality of that complex may be restricted to information that is closely associated with the original threatening experience.
Thrasher, Dalgleish, and Yule (1994) investigated affective Stroop interference in ferry disaster survivors with PTSD. They assessed 13 survivors with PTSD (7 men, 6 women), 20 survivors with scores of less than 40 on the Impact of Events Scale (15 men, 5 women), and 12 non-traumatised subjects; all subjects were matched for age and premorbid IQ (assessed with the National Adult Reading Test). In designing the modified Stroop task, Thrasher, Dalgleish, and Yule (1994) controlled for word length and frequency and emotionality and semantic relations among five sets of 20 words presented in four colours (red, blue, yellow, and green) on cards in the following order: semantically unrelated neutral words, semantically related neutral words (musical instruments), positive emotional words, threat words, and disaster specific words. In general, disaster victims with PTSD were slower to respond than disaster victims without PTSD, who were slower than non-traumatised subjects. In particular, relative to responses to semantically related neutral words, disaster victims with PTSD responded: (a) slower to disaster words than all other subjects, (b) slower to threatening words than the non-traumatised subjects, and (c) normally to positive words. Also, relative to threatening words, disaster victims with PTSD responded slower to disaster words than all other subjects. These results support their hypotheses that disaster victims with PTSD, as opposed to disaster victims without PTSD and non-traumatised subjects, are more sensitive to semantic interference from disaster specific information than generally threatening information, although they are also sensitive to the latter information.
Zeitlin and McNally (1991) assessed implicit and explicit memory for threat-related information in PTSD. They hypothesised that PTSD patients have a threat-related schemata that consists of highly activated elements, such that implicit and explicit memory for threat-related information will be greater in PTSD. They evaluated Vietnam combat veterans with (n = 24) and without (n = 24) PTSD. Subjects were not matched for age, education, or combat exposure, but there was no significant influence of age or education on recall. They were presented with a list of words, consisting of words related to combat, social threat, and positive and neutral affect. Half of the subjects were required to rate how much they liked each word (elaborative encoding) and half of the subjects were required to count the letters in each word (non-elaborative encoding). After counting backwards, subjects were required to complete a word stem completion task with the first word that comes to mind (implicit memory) and a cued recall task with a word that was seen earlier (explicit memory). Half of the words that comprised the word stems in the implicit and explicit memory tasks were presented during encoding, thus memory performance could be assessed for both primed and unprimed words. Combat veterans with PTSD recalled fewer neutral and positive affective words than combat veterans without PTSD. This may reflect a relative paucity of cognitive resources to deal with such information, since most cognitive resources are utilised to process or avoid processing threatening information (Zeitlin and McNally, 1991). This information processing bias was reflected in recall; relative to recall for neutral words, combat veterans with PTSD recalled more combat words than combat veterans without PTSD. However, recall for combat words was significantly associated with combat exposure. Since combat veterans with PTSD had more combat exposure than combat veterans without PTSD, this recall bias may be attributed to greater combat exposure or PTSD. However, combat veterans with PTSD had an implicit memory bias for unprimed combat words, which was enhanced by priming. This implicit memory bias was significantly associated with the severity of PTSD after controlling the degree of combat exposure. Thus, an implicit bias for combat related information in combat veterans with PTSD may reflect a chronic activation of elaborate and stable sensory, perceptual or semantic memories for combat related experiences in combat veterans with PTSD. The activation of these unconscious representations will easily promote conscious awareness of the information contained in them. Thus, the intrusions of combat veterans with PTSD may result from a heightened activation or sensitisation of unconscious combat related memories.
McNally, Litz, Prassas, Shin, Weathers (1994) evaluated episodic memory for personal experience of positive and negative affective states in Vietnam combat veterans with PTSD (n = 39), veterans with other psychiatric conditions (n = 20), and veterans without morbidity (n = 23). Veterans were less intelligent than normal subjects and veterans with PTSD had higher levels of depression, anxiety, and combat exposure than other subjects. Although other studies have reported that demographic characteristics have no significant influence on information processing in PTSD, the demographic characteristics of subjects were not reported, so the results are tentative. Subjects were presented with a short series of pictures that depicted either furniture or combat in Vietnam; furniture was accompanied by classical piano music and combat was accompanied by a variety of combat sounds. Subjects rated their moods before and after the presentations. After the pictures, subjects were required to recall a specific episode in their lives when they experienced several affective states, including positive (comradeship, humor, devotion, gaiety, intimate, kindness, happiness, loyalty, affection, and bravery), negative (boredom, sickness, anxiety, misery, fatigue, shame, panic, sadness, ignorance, and hostility), and neutral (mathematics, amazement, background, determination, reflection, illusion, hierarchy, agreement, gravity, and appearance) affective states. It was expected that the emotional numbing characteristic of PTSD would be related to difficulty recalling specific episodes of emotional experience. The initial pictures affected all subjects equally; relative to furniture, pictures of Vietnam combat reduced happy, positive mood states and increased anxiety, anger, emotional arousal, and negative mood states. Among veterans with PTSD, veterans who were exposed to combat pictures recalled more specific instances of neutral affect, but not positive or negative affect, than veterans exposed to pictures of furniture. Veterans without morbidity reported more specific episodes of positive or negative affect than neutral affect, veterans with other psychiatric conditions reported more specific experiences of negative affect than neutral affect, and veterans with PTSD recalled more specific experiences of negative affect than positive affect. In general, veterans with PTSD were less specific than veterans without morbidity and they recalled more negative memories and memories related to the Vietnam war than veterans with other psychiatric conditions. These tentative results indicate that veterans with PTSD have some difficulty recalling specific emotional experiences, but when they do recall emotional experiences they are more often about negative emotions or associated with their traumatic experiences. This relative deficit in conscious recollection of positive emotional experiences is not merely a result of the impact of their traumatic experiences, since veterans without any morbidity do not have this difficulty. This bias in recollection may be responsible for the onset and maintenance of symptoms of PTSD, so any concerted effort to increase the recall of positive personal experience may counteract the bias and alleviate symptoms of PTSD (McNally et al, 1994).
McNally, Lasko, Macklin, and Pitman (1995) have also investigated explicit memory for experiences of positive and negative affect in combat-related PTSD. They assessed Vietnam combat veterans with (n = 19) and without (n = 13) PTSD. Veterans with PTSD were younger, had less intelligence, and experienced more combat than veterans without PTSD, so their results are only tentative. Veterans were presented with a series of words describing positive and negative personal traits (e.g., friendly, loyal, selfish, cowardly, etc.). They were asked to report the most recent, specific instance or situation they can remember when they were feeling the same way. Veterans with PTSD were generally less specific in their recall of personal experience than veterans without PTSD, especially for positive personal traits. This relative deficiency in the recall of specific instances of personal attributes could be largely attributed to veterans with PTSD who wore regalia of the war that symbolise their role in events that occurred thirty years ago. These veterans had a low proportion of recent specific memories and a high proportion of specific memories related to the war and they recalled only half as many specific memories for negative and positive personal traits as other veterans, which they were much slower to recall. Veterans with PTSD did not differentiate between positive and negative personal traits in either the specificity or speed of recall, whereas veterans without PTSD had more specific memories for positive rather than negative personal traits and they recalled episodes of positive personal traits quicker than negative personal traits. Also, veterans with PTSD were slower to recall a specific instance when they felt positive personal traits than veterans without PTSD. Thus, veterans with PTSD have a relative deficit in their focus of attention or episodic memory for specific experiences of positive feelings about themselves. It is common for people to focus on positive personal attributes more so than negative personal attributes, but veterans with PTSD have no such bias. This relative deficit could be related to disturbances in self-representation and affect, including low self-efficacy, guilt, sorrow, and numbing of positive emotional experience (McNally, Lasko, Macklin, and Pitman, 1995).
It has been postulated that repetitive traumatic intrusions are part of an evolving process of evoking the overwhelming experience of severe trauma so that meaning and purpose can be extracted from the experience (Brett and Ostroff, 1985). A catastrophic experience conflicts with beliefs of safety or invulnerability and can prompt a reassessment of personal security and the risks of danger in the environment (Brett and Ostroff, 1985). The cognitive repetition of the incident may help to extract information from it about exactly what happened and why it happened so that this information can be incorporated into a world view that will shape expectations and values (Brett and Ostroff, 1985).
This information is particularly important in the determination of causal attributions. Perceptions of the relationship between actions and outcomes forms a great part of our self identity and perceived role in life. PTSD patients are often prone to maladaptive causal attributions. They attribute the causes of negative events to external, stable, and uncontrollable sources (Mikulincer and Solomon, 1988). The lack of perceived control of negative life events promotes helplessness (Mikulincer and Solomon, 1988). PTSD patients also attribute the causes of positive events to external and uncontrollable events; they believe that positive experiences are not a result of their own action and therefore derive no pride or self-esteem from achievements (Mikulincer and Solomon, 1988). PTSD patients deny any relationship between their actions and the environment. Coupled with perceptions of the environment that are fraught with fear and expectations of further uncontrollable danger, their hope and desire to act positively would be markedly diminished. In this regard, it is understandable that stressful interpersonal situations provoke an emotion focused problem solving strategy rather than a more effective problem focused strategy in PTSD patients (Nezu and Carnevale, 1987). Rather than acknowledge that a problem exists and that their actions could remedy the situtation, PTSD patients focus on their feelings of distress and withdraw into their feelings, away from the action(s) that could resolve their problem.
Bryant, R., Harvey, A. (1995). Processing threatening information in posttraumatic stress disorder. Journal of Abnormal Psychology, 104: 537-541.
Cassiday, K., McNally, R., Zeitlin, S. (1992). Cognitive processing of trauma cues in rape victims with post-traumatic stress disorder. Cognitive Therapy and Research, 16: 283-295.
Chemtob, C., Roitblat, H., Hamada, R., Carlson, J., Twentyman, C. (1988). A cognitive action theory of post-traumatic stress disorder. Journal of Anxiety Disorders, 2: 253-275.
Foa, E., Feske, U., Murdock, T., Kozak, M., McCarthy, P. (1991). Processing of threat-related information in rape victims. Journal of Abnormal Psychology, 100: 156-162.
Foa, E., Steketee, G., Olasov-Rothbaum, B. (1989). Behavioral/Cognitive conceptualisations of post-traumatic stress disorder. Behaviour Therapy, 20: 155-176.
Goenjian, A., Najarian, L., Pynoos, R., Steinberg, A., Manoukian, G., Tavosian, A., Fairbanks, L. (1994). Posttraumatic stress disorder in elderly and younger adults after the 1988 earthquake in Armenia. American Journal of Psychiatry, 151: 895-901.
Koopman, C., Classen, C., Spiegel, D. (1994). Predictors of posttraumatic stress symptoms among survivors of the Oakland/Berkeley, Calif., Firestorm. American Journal of Psychiatry, 151: 888-894.
Litz, B., Keane, T. (1989). Information processing in anxiety disorders: application to the understanding of post-traumatic stress disorder. Clinical Psychology Review, 9: 243-257.
McCarroll, J., Ursano, R., Fullerton, C. (1993). Symptoms of posttraumatic stress disorder following recovery of war dead. American Journal of Psychiatry, 150: 1875-1877.
McCarroll, J., Ursano, R., Fullerton, C. (1995). Symptoms of PTSD following recovery of war dead: 13-15 month follow-up. American Journal of Psychiatry, 152: 939-941.
McNally, R., Kaspi, S., Riemann, B., Zeitlin, S. (1990). Selective processing of threat cues in posttraumatic stress disorder. Journal of Abnormal Psychology, 99: 398-402.
McNally, R., Lasko, N., Macklin, M., Pitman, R. (1995). Autobiographical memory disturbance in combat-related posttraumatic stress disorder. Behaviour Research and Therapy, 33: 619-630.
McNally, R., Litz, B., Prassas, A., Shin, L., Weathers, F. (1994). Emotional priming of autobiographical memory in post-traumatic stress disorder. Cognition and Emotion, 8: 351-367.
McNally, R., Shin, L. (1995). Association of intelligence with severity of posttraumatic stress disorder symptoms in Vietnam combat veterans. American Journal of Psychiatry, 152: 936-938.
Schnurr, P., Friedman, M., Rosenberg, S. (1993). Premilitary MMPI scores as predictors of combat-related PTSD symptoms. American Journal of Psychiatry, 150: 479-483.
Thrasher, S., Dalgleish, T., Yule, W. (1994). Information processing in post-traumatic stress disorder. Behaviour Research and Therapy, 32: 247-254.
Trandel, D., McNally, R. (1987). Perception of threat cues in post-traumatic stress disorder: semantic processing without awareness? Behaviour Research and Therapy, 25: 469-476.
Warshaw, M., Fierman, E., Pratt, L., Hunt, M., Yonkers, K., Massion, A., Keller, M. (1993). Quality of life and dissociation in anxiety disorder patients with histories of trauma or PTSD. American Journal of Psychiatry, 150: 1512-1516.
Weine, S., Becker, D., McGlashan, T., Laub, D., Lazrove, S., Vojvoda, D., Hyman, L. (1995). Psychiatric consequences of "ethnic cleansing": Clinical assessments and trauma testimonies of newly resettled bosnian refugees. American Journal of Psychiatry, 152: 536-542.
Yehuda, R., Kahana, B., Schmeidler, J., Southwick, S., Wilson, S., Giller, E. (1995). Impact of cumulative lifetime trauma and recent stress on current posttraumatic stress disorder symptoms in holocaust survivors. American Journal of Psychiatry, 152: 1815-1818.
Yehuda, R., McFarlane, A. (1995). Conflict between current knowledge about posttraumatic stress disorder and its original conceptual basis. American Journal of Psychiatry, 152: 1705-1713.
van der Kolk, B., Herron, N., Hostetler, A. (1994). The history of trauma in psychiatry. Psychiatric Clinics of North America, 17: 583-600.
Zeitlin, S., McNally, R. (1991). Implicit and explicit memory bias for threat in post-traumatic stress disorder. Behaviour Research and Therapy, 29: 451-457.