Three concepts are important: (1) we all know more than we can say, (2) memory is divided into two aspects: implicit and explicit, and (3) memory is not simply a computer-like recollection of past events but is partly constructed over time.

1. We all know more than we can say.

 

The first, that we know more than we can say, lies at the heart of core cognitive schemata. Our early (and sometimes later) experiences lay down the framework for a series of tacit rules and assumptions about the way the worlds work. For example, a child who is faced with inconsistent and punitive parenting may develop an unspoken rule that, "People can't be trusted". Or, a child deprived of needed resources may develop an unspoken rule that, "I must fight for everything I get." A child who is rejected by primary care givers (most often parents) may develop a rule that, "I'm unlovable". These are examples of Beck's core cognitive schemata. However, the very fact that they were developed early, often before language development, means they may not be easily identified.

 

2. Memory is divided into two aspects: implicit and explicit.

 

Explicit memory is invoked when the individual recalls an event in the past. In this case, memory is used as an object. Explicit memory requires a conscious and intentional recall and is often slow and uncertain in its final form. Implicit memory, on the other hand, occurs when we use our past knowledge (for example in problem solving) without conscious awareness that we are using it. Implicit learning occurs through the tacit detection of covariation (things that occur together).

Memory in this case is used as a tool and resembles in some way what is commonly called the unconscious, through without the extraneous conceptual baggage contained in that term. Not only is there a sharp dissociation between explicit and implicit memory, but the latter tends to be richer in content, to occur earlier in the evolutionary process, to occur faster, and to be more robust in nature than the former.

 

The implicit knowledge that forms core cognitive schemata acts like a template or mental filter, constraining what can enter the cognitive system as new data. Thus, the implicit rule that, "People can't be trusted" acts as a filter to incorporate examples of other people's untrustworthiness while excluding examples of their trustworthiness. In other words, we notice those events that invalidate it. This is known as confirmatory bias and is part of the cognitive processing mechanisms of all humans. It results in the truism that we see/find what we expect to see/find and requires a great effort or will to overcome.

 

3. Memory is not simply a computer-like recollection of past events but is partly constructed over time.

Contrary to popular thought, the mind does not act like a computer and store data in pure form to be retrieved later unchanged. The remembrance of past events are shaped by wishes, phantasy, and the like. All memories undergo changes over time. The amount of change in the subjective experience of memory seems to depend on three variables: context, attention and number of times an episode is recalled.

Context refers to aspects of the environment as well as one's inner state. It has been documented that memory for an event improves when the individual is in a similar context to that in which the memory was originally learned. For example, sad or angry clients often spontaneously remember past events when they felt the same emotion.

In order to process information into memory, people must pay attention to it. However, there are degrees of attention and attention is often divided. People learn differently in focused versus divided attention situations. Information learned in divided attention cannot be recalled as easily as that learned in focused attention but individuals still may be able to recognize it or assess familiarity with it. They may also be able to use the information though they may not be aware that they are doing so nor be able to monitor its use. The hyponotherapist, Milton Ericson, often used divided attention to help clients incorporate new data or to process old data differently. The divided attention prevented the client from consciously checking on why something came to mind and applying other cognitive censoring mechanisms.

The number of times an event is recalled has a powerful impact on memory though not in the way it is commonly assumed. Research has indicated that events recalled repeatedly become less accurate but are held with more confidence. Thus, individuals who repeatedly tell the same story become more sure of it over time but tell it less accurately, a phenomenon known to anyone who has had to listen to an old veteran's war stories. Thus, memory for an event is gradually changed over time, often in a self-serving direction. Marriage and family therapists are often astonished at the wide difference among family members in recalling past events, all of which are held with the utmost sincerity and vigor.

Exacerbating this problem in the tacit detection of covariation mentioned earlier, as the basis of implicit learning. Essentially, this means that things that occur together, however accidentally, will be remembered as related - and it will often be assumed that one caused the other. This is the basis of the logical fallacy, ad hoc ergo propter hoc (after this, therefore because of this), as well as the associational learning of behaviorism and is both an advantage and a disadvantage. Humans have an amazing ability to detect subtle covariations that are often advantageous to survival but, precisely because of this ability, they may also detect covariations that don't really exist. This is especially true in ambiguous situations, where people tend to impose pre-existing cognitive templates on the situation and thus interpret it on the basis of past tacit rules and assumptions. Because these rules were learned implicitly, they are very resistant to conscious awareness and subsequent change.

These human cognitive tendencies pose a challenge for the cognitive therapist attempting to help clients uncover past events or core cognitive schemata. In doing so, especially if therapists are guided by a powerful and deeply-held belief system of human pathology, there is a real danger that they will tacitly "encourage" clients to shape their memories according to the therapist's theoretical outlook or even "remember" events that never happened. Therapists may also "notice" connections (covariations) where non "really" exist. Therapists as disparate as Fiedler and Meichenbaum have noted that more successful clients tend to adopt the ideas of their therapists (Fiedler, 1950; Meichenbaum, 1977). While in many situations, the clients subjective memory (regardless of its accuracy or "truth") is all that counts, in other situations (particularly legal) more objective accuracy is critical. Therapists should always treat their most cherished ideas as working hypotheses, be aware of the confirmatory bias tendency, and seek independent confirmation supporting the memory of important events.

Therefore, cognitive therapists should adopt the true scientific method (though it is actually practices by few scientists). There is powerful human tendency to develop hypothesis (formed in part by our tacit cognitive structure) and then to seek confirmation of these hypothesis. Researches do this and so do clients. A better strategy, however, may be to vigorously seek disconfirming data of our hypothesis. In this model, the attempt is not to prove our hypothesis but to disprove them. While it's a more difficult (and humbling) attitude to practice, it may be more beneficial in the long run.

Beck, AT, Freeman, A., Pretzer, J., Davis, D., Fleming, B., Ottaviani, R., Beck, J., Simon, K., Padesky, C., Meyer, J., & Trexler, L. (1990). Cognitive therapy of personality disorders. New York, Guilford Press.

Dowd, E.T. & Courchaine, K.E. (1996) Implicit learning, tacit knowledge and implication for stasis and change in cognitive psychotherapy. Journal of Cognitive Psychotherapy, 10, 163-180.

Fiedler, F.E. (1950) A comparison of therapeutic relationships in psychoanalytic, non-directive and Adlerian therapy. Journal of Consulting Psychology, 14, 436-445.

Meichenbaum, D. (1977). Cognitive-behavior modification: An integrative approach. New York, Plenum Press