This paper explores the nature of hippocampal damage in post-traumatic stress disorder (PTSD) and the implications of this damage for cognitive functions in PTSD. Several reports now demonstrate reduced volume of hippocampus in PTSD and various animal studies implicate stress-related glucocorticoids in the atrophy of hippocampal CA3 pyramidal cells. Functional neuroimaging of the hippocampus now clearly demonstrates that it performs a vital role in detection of novelty, episodic memory consolidation and recall, and a behavioural inhibition system. Thus, the cognitive implications of damage to the hippocampus in PTSD are an impairment of episodic memory and novelty detection that gives rise to disruption of executive functions and results in uncertainty, distraction, and anxiety.
Bremner et al (1995) found an 8% reduction of right hippocampal volume in Vietnam veterans with PTSD, which was associated with deficits in verbal short-term memory (WMS).
Gurvitis et al (1996) report bilateral hippocampus reduction in combat PTSD, after controlling for age and whole brain volume. The decrease was greater in left (26%) than right (22%) hippocampus, especially after controlling for excessive alcohol consumption and combat exposure. Hippocampal volume was strongly related to both combat exposure and PTSD symptom severity. Furthermore, hippocampal volume correlated with several measures of attention and memory (digit span, arithemetic [WAIS-R], attn/conc. index [WMS], and delayed recall errors [Benton visual retention test]). There were no differences in intracranial cavity, whole brain, ventricles, ventricle:brain ratio, or amygdala (although, the right amygdala was larger in patients than controls, which approached significance at .07).
Bremner et al (1997) report a 12% reduction of left hippocampus in adult survivors of chronic childhood abuse (7 to 15 years of physical/sexual/emotional abuse). 10% of this reduction was solely related to PTSD diagnosis, when sex, age, education or alocohol abuse were controlled in linear regression. There was a 5% reduction in right hippocampus that was not significant and there were no differences in caudate or amygdala volume, but there was a larger left temporal lobe in patients than controls. Reduced left hippocampal volume was not correlated with significant verbal memory impairments (immediate and delayed recall and retention), trauma onset or duration, years since trauma cessation, or PTSD symptom severity. Trauma early in life may damage hippocampal structures, but neural plasticity could adapt to this damage and allocate verbal memory processes to other structures, thereby diminishing the relationship between hippocampal volume and verbal memory impairment. Larger left temporal lobes could be related to better visual performance in these patients, which might be an adaptation to a loss of verbal capacity.
Stein et al (1997) also reported a 5% reduction of left hippocampus in women survivors of severe childhood sexual abuse.
Canive et al (1997) report focal white matter lesions in 8 of 42 combat related PTSD patients using a FLAIR MRI acquisition. The lesions were in periventricular regions or near the white/grey cortical junctions. The lesions were not associated with symptom severity or comorbid depression or alcohol abuse.
Press, Amaral, and Squire (1989) report a 49% reduction of hippocampal volume in amnesic patients, with virtually normal parahippocampal volumes. These patients showed severe deficits of verbal and non-verbal memory, while performing normally on various other tasks (Press et al, 1989).
The CA3 region of the hippocampus is a major target for glucocorticoids in the brain, as it plays an important role in regulating the pituitary-adrenocortical response to stress. Increased receptivity of the hippocampus or hypothalamus could be an adaptation to increased stress related glucocorticoid levels and thereby diminishes subsequent cortisol responses to stress (Murburg, 1997). However, excess and chronic glucocorticoid levels induce atrophy and dendritic loss in the hippocampus by disrupting cellular metabolism and increasing cellular susceptibility to excitatory amino acids such as glutamate (McEwen, 1999; McEwen and Magarinos, 1997; Sapolsky, 1996; Watanabe, Gould, and McEwen, 1992c). Inhibition of glucocorticoid stimulation by tianeptine and phenytoin, reduces stress related hippocampal atrophy (McEwen, 1999; Watanabe et al 1992a,b).
Glucocorticoids damage CA3 in the hippocampus – what functions does this region perform and how is it related to other subfields of the hippocampus?
The anterior CA1 region of the hippocampus receives inputs from both sensory/perceptual systems of the temporal, parietal, and olfactory cortices, which generate multimodal or contextual representations of the behavioural environment, and executive systems of the parietal and prefrontal cortices, which integrate sensory representations with movement intentions (Eichenbaum and Otto, 1993). Coherent interactions of temporal regions involved in object perceptual processing and parietal regions involved in spatial orientation and motor guidance have been implicated in the Hoffding step, the moment of perception (Dolan et al, 1997; Rodriguez et al, 1999; von Stein, Rappelsberger, Sarnthein, and Petsche, 1999). Activity from these temporal and parietal regions has been found to project directly to CA1 pyramidal cells of the hippocampus, suggesting that the hippocampus directly recieves multimodal representations of stimulus information (Rockland and Hoesen, 1999).
The multimodal inputs generated by the neocortex contain the highest level of representation of the stimulus environment and movement intentions (Eichenbaum and Otto, 1993). Various evidence clearly indicates that hippocampal cells respond to various spatial and object properties of the stimulus environment as well as the position or intention attributes of the organism in the environment (Eichenbaum and Otto, 1993). For instance, some cells of anterior hippocampal regions, CA1 and CA3, are involved in allocentric spatial mapping of objects into episodic memories (ie, independent of the position or orientation of the observer), but these cells interact with and rely on neighbouring cells that spatially map head position/orientation and relative position of the self with regard to these objects (Georges-Francois, Rolls, and Robertson, 1999; Rolls, 1996). Similarly, the left hippocampus is responsive to verbal or linguistic representations (Wagner et al, 1998), wheras bilateral or right hippocampus is responsive to pictorial or object representations (Brewer et al, 1998). Moreover, the hippocampus is responsive to the higher order contextual relevance or significance of stimuli. For instance, anterior hippocampal regions respond to perceptual novelty, whereas posterior regions respond to stimuli that have response related familiarity (Strange et al, 1999).
It is generally accepted that the hippocampus is involved in the generation and recollection of episodic memories. However, not all experiences are remembered – it is mostly experiences that contribute to our understanding of the world that are remembered. The role of the hippocampus is to detect or extract the signficant information for further memory consolidation from the less important, repetitive material that is already remembered. The hippocampus is involved in the ongoing comparison of cognitive expectations against reality for the purpose of maintaining accurate representation of reality that will aid future expectations (eg, Strange et al, 1999). In performing this process, the hippocampus constructs higher order or supramodal episodic representations that contain both stimulus and response information over time (Eichenbaum and Otto, 1993; Martin, Wiggs, Ungerleider & Haxby, 1996).
Two important aspects of cognition influence the effectiveness of memory encoding: (a) undivided attention and (b) encoding the meaning rather than stimulus properties (Rugg, 1998). The first of these, attention, is either consciously directed or engaged by novelty. Once attention is directed to a percept, it is likely that evaluation processes translate the simple sensory information into meaningful representations, which facilitate encoding (Rugg, 1998). Thus, novelty is an important determinant of what is encoded into episodic memory (Tulving and Kroll, 1995).
The hippocampus is a key structure in the detection of novelty or familiarity. Evidence indicates that activation along the anterior-posterior axis of the hippocampus reflects a distribution of stimulus novelty to familiarity. For instance, Strange et al (1999) report that anterior hippocampal regions respond to perceptual novelty, whereas posterior regions respond to stimuli that have response related familiarity. They propose that the anterior hippocampus registers mismatches of current stimulus information against an expected or predictive representation. When mismatches are identified and they have behavioural significance, greater attention and familiarity with the information engages posterior hippocampal regions. Furthermore, the activity along this hippocampal axis for the detection of novelty or familiarity may help to explain anterograde or retrograde amnesia. Strange et al (1999) postulate that damage to the anterior hippocampus impairs novelty detection and episodic memory consolidation, which explains anterograde amnesia. Conversely, damage of the posterior hippocampus impairs recall of past experience, which explains retrograde amnesia (Strange et al, 1999).
There is a case to be made that hippocampal and parahippocampal regions are involved in detection of novelty, which subseqently or concurrently initiates more complex encoding strategies in frontal regions. The reports of Wagner et al (1998) and Brewer et al (1998) indicate that left prefrontal and parahippocampal regions are active during encoding and consolidation in an incidental memory task. Their task materials did not involve novelty and did not elicit activation in the hippocampus proper. It may be that the hippocampus is not directly involved in encoding or consolidation, but merely detects novelty and thereby initiates more complex evaluation, encoding, and consolidation processes. This interpretation is supported by a dissociation of frontal and hippocampal functions reported by Dolan and Fletcher (1997). They found that the left prefrontal cortex is engaged by changes in the content of linguistic category-exemplar encoding processes, implicating this brain area in the controlled transformation of auditory-verbal stimuli into meaningful information. On the other hand, left medial temporal lobe areas, including the hippocampus, were responsive to contextual novelty. The physical properties of the auditory-verbal stimuli were familiar; the novelty of the stimuli was related to the encoded associations of the verbal information – the medial temporal lobe structures responded to the associative or contextual novelty of the material, not simply the physical attributes of the stimuli (Dolan and Fletcher, 1997). The hippocampal detection of novelty may initiate frontal executive processes that alter encoding strategies.
See also Squire and Zola (1996) – relevant to this section.
A wide variety of evidence now indicates that the hippocampus provides an episodic or contextual framework for multimodal or semantic representations that are encoded in the interactions of the parahippocampal and association cortices (Eichenbaum, 1997). Semantic memories are encoded and consolidated by interactions of parahippocampal and associative or executive cortex, in the absence of hippocampal input (Eichenbaum, 1997; Desimone, 1996; Brewer et al, 1998; Wagner et al, 1998). Various studies report that such memories are established in animals and humans with hippocampal damage, but not so with parahippocampal damage (Eichenbaum, 1997). Furthermore, evidence that the hippocampus provides contextual or episodic information arises from deficits in locating objects or faces in their spatial context among individuals with developmental hippocampal pathology (Vargha-Khadem et al, 1997).
The contextual or episodic processes of the hippocampus modify the multimodal or semantic representations in two important ways (Eichenbaum, 1997). Firstly, it assists the identification of an object or event in its current or previous context. Secondly, it can differentiate an object or event from its context, which is important for perceptual constancy. For instance, after initially meeting a person in one context, we need to be able to recognise them again a day later while they are wearing different clothes in a different building. This requires considerable extraction of person specific percepts from a potentially vast array of contextual information.
Furthermore, Eichenbaum (1997) proposes that the hippocampal inputs into associative memories can influence the latter retrieval cues for the memories. The quality or amount of episodic information and the manner of integration of that information into the associative memory networks will affect what type and how many cues can elicit memories and how much of the memories they elicit. For instance, an interleaved encoding of episodic with various multimodal or semantic representations can better facilitate retrieval of the whole episode and it’s specific representations (Eichenbaum, 1997).
In general, the interactions of the cortex and parahippocampal structures serve to integrate and hold multimodal and semantic representations, while the hippocampus serves as a comparator and organiser of these representations (Dusek & Eichenbaum, 1997; Eichenbaum, Schoenbaum, Young & Bunsey, 1996; Goldman-Rakic, 1996). An insight into this relationship between the hippocampus and the parahippocampal region has been recently provided by Iijima et al (1996). They propose that reverberating circuits in the entorhinal cortex (EC) and cyclic interactions between the EC and hippocampus play an important role in holding information and selectively gating it into hippocampal circuits. The superficial layers of the EC receive multimodal input from association cortices. The EC can selectively gate this information into a spatial and temporal integration in reverberating circuits that hold the information. Furthermore, the EC is a major cortical input into the hippocampus and it can selectively gate the information from its reverberating circuits into the hippocampus. The hippocampus can, in turn, process this information and return inputs from CA1 and the subiculum into the deep layers of the EC, where it can reintegrate into the EC circuits or feedback into the association cortex. Iijima et al (1996) demonstrated, in real-time, that the reverb ciruit of the rat EC oscillates at about 5 Hz, which included activation and feeback in the hippocampus (see http://science-mag.aaas.org/science/feature/beyond/#iijima).
Memory consolidation is generally considered a process of long-term potentiation (LTP) of nerve cell membranes. Although the precise neurochemisty is not known, there are a variety of good models proposed (see Lisman and Fallon, 1999). Several models of early phases of LTP implicate membrane proteins and second messenger systems in positive feedback processes designed to maintain an active membrane state (Lisman and Fallon, 1999). Models of later phases of LTP implicate gene expression in stable changes of membrane or synapse structure and functions (Lisman and Fallon, 1999).
For neurochemistry, see also Kandel (1996).
Bear (1996) reports that LTP and long-term depression occur in both the hippocampus and the cerebral cortex.
The hippocampus not only signals novelty in its role in episodic memory, it also acts as a warning system for behaviour when significant mismatches between expectations and reality occur. This alterting system is what Gray (1988, 1982a,b) has called the behavioural inhibition system (BIS), which plays an important part in anxiety. The BIS is activated by unpredictable stimuli that require reappraisal of appropriate action plans. The anterior hippocampus is involved in registration of mismatches between expectations based on recent experience and current events (Gray, 1988, 1982a,b; Strange et al, 1999). For instance, Strange et al (1999) found that anterior hippocampal regions respond to perceptual and response related novelty, whereas posterior regions respond to familiar stimuli that have response implications or meaning. The latter activation could be involved in retrieval of information for further stimulus comparisons or the consolidation of appropriate stimulus-response associations.
What is the relationship between novelty and emotion, especially anxiety? How is the hippocampal novelty detection process related to functions of the amygdala? - see McGaugh (1996), who proposes that the effects of stress hormones on memory are mediated through the amygdala – see LeDoux etc.
Damage to the anterior portion of the hippocampus in PTSD can affect the ability to accurately develop supramodal, episodic representations and accurately evaluate and incorporate new information into episodic memory. This lack of accuracy in episodic memory will continually generate mismatch activity in the hippocampus that will initially hyperarouse the BIS and initiate excess stress reactions. These responses may attempt to habituate, but cannot do so due to impaired episodic memory evaluation and consolidation, which leads to cognitive, emotional, and endocrine exhaustion.
In this manner, PTSD patients are unable to incorporate new information after their trauma and the most recent episodic memories that were consolidated remain traumatic and continue to affect expectations of the world. This fundamentally alters their world view and psychic life.
Threshold of hippocampal activation for behavioural inhibition system is reached by threatening information and this information is incorporated into episodic memory and generates further traumatic associations that encourage intrusions, but neutral information doesn’t reach this threshold and so doesn’t get properly evaluated and integrated into episodic memory. The BIS shifts attention and executive functions away from neutral information toward threat.
Note similarity of this idea to that of Murburg (1997), "relative basal quiescence of [catecholamine] systems with enhanced responsivity to stimulation may provide for enhancement of ‘signal to noise ratios’ in neuronal and other systems. Such enhancement may facilitate selective attention and selective responding to the most strongly determined inputs …, potentially contributing to symptoms including hypervigilance, insomnia, flashbacks, intrusive memories, panic, physiological hyperactivity, and startle."
Murburg (1997) also points out interesting peripheral relationships to central processes. For instance, traumatic visualisations may induce increased peripheral epinephrine that increases memory consolidation by stimuluating the amygdala, which is important in emotional memories and was identified, along with increased activity in anterior cingulate and decreased activity in left inferior frontal cortex, in a rCBF study by Rauch and Shin (1997). Murburg (1997) also notes that, "responding to a stressor … may itself leave behind molecular ‘memory traces’ that so alter involved neural pathways as to predispose them to be more readily activated in the future."
Relationship of hippocampal volume to ERP stimulus evaluation amplitudes - hippocampal atrophy, episodic memory impairment, behavioural inhibition and executive function disruption, all impair integration of new non-threatening information into neutral information schemata. This impairs attentional strategies that are directed by knowledge of what to expect and when and where to expect it. It thereby impairs the ability to detect, evaluate, and consolidate knowledge about the regularities in various, especially complex, stimulus arrays.
Is it possible to administer a pharmacological manipulation of hippocampal novelty detection in PTSD?
Bear, M. (1996). . Proceedings of the National Academy of Science, 93:13453-13459.
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., Randall, P., Vermetten, E., Staib, L., Bronen, R., Mazure, C., Capelli, S., McCarthy, G., Innis, R., Charney, D. (1997). Magnetic resonance imaging-based measurement of hippocampal volume in posttraumatic stress disorder related to childhood physical and sexual abuse: A preliminary report. Biological Psychiatry, 41:23-32.
Brewer, J., Zhao, Z., Desmond, J., Glover, G., Gabrieli, J. (1998). Making memories: Brain activity that predicts how well visual experience will be remembered. Science, 281:1185-1187.
Canive, J., Lewine, J., Orrison, W., Edgar, C., Provencal, S., Davis, J., Paulson, K., Graeber, D., Roberts, B., Escalona, P., Calais, L. (1997). MRI reveals gross structural abnormalities in PTSD. Annals of the New York Academy of Sciences, 821:512-515.
Desimone, R. (1996). . Proceedings of the National Academy of Science, 93:13494-13499.
Dolan, R., Fink, G., Rolls, E., Booth, M., Homes, A., Frackowiak, R., Friston, K. (1997). How the brain learns to see objects and faces in an impoverished context. Nature, 389: 596-599.
Dusek, J., Eichenbaum, H. (1997). The hippocampus and memory for orderly stimulus relations. Proceedings of the National Academy of Science, 94:7109-7114.
Eichenbaum, H. (1997). How does the brain organise memories? Science, 277:330-332.
Eichenbaum, H., Otto, T. (1993). Where perception meets memory: Functional encoding in the hippocampus. In T. Ono, L. Squire, M. Raichle, D. Perrett, M. Fukuda (Eds.), Brain Mechanisms of Perception and Memory: From Neuron to Behaviour. New York: Oxford University Press.
Eichenbaum, H., Schoenbaum, G., Young, B., Bunsey, M. (1996). Functional organisation of the hippocampal memory system. Proceedings of the National Academy of Science, 93:13500-13507.
Georges-Francois, P., Rolls, E., Robertson, R. (1999). Spatial view cells in the primate hippocampus: Allocentric view not head direction or eye position or place. Cerebral Cortex, 9:197-212.
Goldman-Rakic, P. (1996). Memory: Recording experience in cells and circuits: Diversity in memory research. Proceedings of the National Academy of Science, 93:13435-13437.
Gray, J. (1988). The neuropsychological basis of anxiety. In C. Last and M. Hersen (Eds.), Handbook of Anxiety Disorders. New York: Pergammon Press.
Gray, J. (1982a). Precis of the neuropsychology of anxiety: An enquiry into the functions of the septo-hippocampal system. Behavioural and Brain Sciences, 5:469-534.
Gray, J. (1982b). The Neuropsychology of Anxiety: An enquiry into the functions of the septo-hippocampal system. Oxford: Oxford University Press.
Gurvits, T., Shenton, M., Hokama, H., Ohta, H., Lasko, N., Gilbertson, M., Orr, S., Kikinis, R., Jolesz, F., McCarley, R., Pitman, R. (1996). Magnetic resonance imaging study of hippocampal volume in chronic, combat-related posttraumatic stress disorder. Biological Psychiatry, 40:1091-1099.
Iijima, T., Witter, M., Ichikawa, M., Tominaga, T., Kajiwara, R., Matsumoto, G. (1996). Entorhinal-hippocampal interactions revealed by real-time imaging. Science, 272:1176-1179.
Kandel, E. (1996). . Proceedings of the National Academy of Science, 93:13445-13452.
Lisman, J., Fallon, J. (1999). What maintains memories? Science, 283: 339-340.
Martin, A., Wiggs, C., Ungerleider, L., Haxby, J. (1996). Neural correlates of category-specific knowledge. Nature, 379:649-652.
McEwen, B. (1999). Stress and hippocampal plasticity. Annual Review of Neuroscience, 22:105-122.
McEwen, B., Magarinos, A. (1997). Stress effects on morphology and function of the hippocampus. Annals of the New York Academy of Sciences, 821:271-284.
McGaugh, J. (1996). . Proceedings of the National Academy of Science, 93:13508-13514.
Murburg, M. (1997). The psychobiology of post-traumatic stress disorder: An overview. Annals of the New York Academy of Sciences, 821:352-358.
Rockland, K., Hoesen, G. (1999). Some temporal and parietal cortical connections converge in CA1 of the primate hippocampus. Cerebral Cortex, 9:232-237.
Rodriguez, E., George, N., Lachaux, J., Martinerie, J., Renault, B., Varela, F. (1999). Perception’s shadow: long-distance synchronisation of human brain activity. Nature, 397:430-433.
Rolls, E. (1996). A theory of hippocampal function in memory. Hippocampus, 6:601-620.
Rugg, M. (1998). Memories are made of this. Science, 281:1151-1152.
Sapolsky, R. (1996). Why stress is bad for your brain. Science, 273:749-750.
Stein, M., Hanna, C., Koverola, C., Torchia, M., McClarty, B. (1997). Structural brain changes in PTSD: Does trauma alter neuroanatomy? Annals of the New York Academy of Sciences, 821:76-82.
Strange, B., Fletcher, P., Henson, R., Friston, K., Dolan, R. (1999). Segregating the functions of human hippocampus. Proceedings of the National Academy of Sciences, 96:4034-4039.
Squire, L., Zola, S. (1996). Structure and function of declarative and nondeclarative memory systems. Proceedings of the National Academy of Science, 93:13515-13522.
Tulving, E., Kroll, N. (1995). Pychonomic Bull. Review, 2:387-390.
Vargha-Khadem, F., Gadian, D., Watkins, K., Connelly, A., Van Paesschen, W., Mishkin, M. (1997). Differential effects of early hippocampal pathology on episodic and semantic memory. Science, 277:376-380.
von Stein, A., Rappelsberger, P., Sarnthein, J., Petsche, H. (1999). Synchronization between temporal and parietal cortex during multimodal object processing in man. Cerebral Cortex, 9:137-150.
Wagner, A., Schacter, D., Rotte, M., Koutstaal, W., Maril, A., Dale, A., Rosen, B., Buckner, R. (1998). Building memories: Remembering and forgetting of verbal experiences as predicted by brain activity. Science, 281:1188-1191.
Watanabe, Y., Gould, H., Cameron, D., Daniels, D., McEwen, B. (1992a). Phenytoin prevents stress- and corticosterone-induced atrophy of CA3 pyramidal neurons. Hippocampus, 2:431-436.
Watanabe, Y., Gould, H., Daniels, D., Cameron, D., McEwen, B. (1992b). Tianeptine attenuates stress-induced morphological changes in the hippocampus. European Journal of Pharmacology, 222:157-162.
Watanabe, Y., Gould, H., McEwen, B. (1992c). Stress induces atrophy of apical dendrites of hippocampal CA3 pyramidal neurons. Brain Research, 588:341-345.