Managing Memory and Metamemory Impairments in Individuals with Traumatic Brain Injury

Buried in the middle of an e-mail from my sister was news I had hoped never to receive-"By the way, did you hear that Mark fell off his bike, hit his head, and has lots of stitches?"

Mark is my brother-in-law. I immediately switched from the sister role to the role of medical investigator, and peppered her with questions to learn more about the severity of his injury: "Was he wearing his helmet? Did he lose consciousness and if so, for how long? Was he confused? Did the MRI show anything? Has he returned to work?" Mark was lucky-the blow to his head did not result in any lasting cognitive or memory problems. He was back to work in a few days.

Many others are not as fortunate. Every year, about 500,000 civilians sustain a traumatic brain injury (TBI). The Centers for Disease Control and Prevention estimates more than 50 million people currently live with TBI-related disabilities in the United States. During times of war, this number swells. Unfortunately, no age group is spared although young men are more likely to be injured than women or members of other age groups.

A blow to the head can result in long-lasting motor, cognitive, communicative, and psychosocial disabilities. These disabilities range from mild to severe and act as significant barriers when trying to return home, to school, and to work. Speech-language pathologists are uniquely trained to manage these cognitive-communication disorders, as ASHA technical documents attest.

Unlike the common portrayal of memory loss in feature films, memory impairment after TBI typically reflects difficulty individuals have remembering recent events, not their identity or remote past. While some individuals with severe injuries may have forgotten parts of their past-called retrograde amnesia-it is much more common to have trouble remembering recent events or details, called anterograde amnesia. During the early stages of recovery when individuals have fleeting attention skills, it is quite common for individuals to be very confused. This confusion presents as impaired comprehension, inability to follow instructions, disorganized discourse, confabulation, word substitutions, and even disrupted phonology. As attention improves, confusion subsides and individuals begin to form new memories, i.e., begin to learn.

Even after the "acute" recovery phase however, most individuals continue to be forgetful. They don't remember the details from stories or conversation, they have trouble remembering appointments the next day or tasks they need to plan for, they have difficulty remembering people's names, and they struggle to recall changes in procedures at work. These short-term memory problems are related to the injury.

The hippocampus, which is located just inside the temporal lobes, is where short, brief memories are formed and briefly stored while actively "waiting" to associate with other bits of information. The temporal and frontal lobes rest on the skull's boney shelf. Together, these lobes and their connections to other brain regions often are injured from direct or indirect blows to the head. Fortunately, injured individuals have other types of memory that remain intact, such as memory for procedures and routines. In treatment, SLPs should capitalize on these procedural memory skills when training the compensatory strategies.

Strategies

Providing individuals with strategies does not guarantee that they will select the appropriate strategy for a given task or use it at the right time. What a person "does with what they have is more important than what they actually have" (Ylvisaker & Feeney, 2004). To make strategy decisions, individuals need to acknowledge or "know" that they have memory impairment and therefore have a need for the strategy.

Metamemory is thinking about your memory. "Meta" refers to one's ability to view, observe, and assess more basic cognitive processes…" (Kennedy & Coelho, 2005, p. 243). "Meta" can be applied to other systems, as in metacognition, metacomprehension, or metalinguistics. Strong evidence from many disciplines links meta processes to frontal lobe activity. Unfortunately, the areas of the brain that allow us to be "aware" are areas likely to be injured in a TBI. Thus, many individuals with TBI have dual disabilities-a memory impairment and metamemory impairment. 

Three aspects of a meta system are important to understand, especially for SLPs, neuropsychologists, and other rehabilitation professionals working with individuals with brain injury. These aspects include autobiographical beliefs (also called self-awareness), self-monitoring, and self-control during activities. Autobiographical beliefs are opinions one holds about general skills. They are updated by ongoing experiences over a period of time. When recovering from TBI, most individuals become aware of physical impairments before gaining awareness of more subtle impairments such as attention and memory. The longer one lives with the disabilities associated with TBI, the more aware they become. But of course, there is tremendous heterogeneity in this population; some remain "unaware" for an extended time, whereas others' awareness emerges as they recover and stabilize over months and years. 

Updating self-awareness of memory is related to memory for recent events. We need to be able to remember daily memory experiences (particularly memory failures) in order to reflect on these and update our general belief about our memory. For example, if people remember that they forgot about an appointment and missed it, they are more likely to think that their memory is not as good as it used to be. Thus, sufficient memory for daily events is necessary if individuals are going to become more realistic about their memory over time.

When individuals are accurate in predicting their memory, they are more likely to decide to use a strategy that could help them remember. These strategy decisions are called self-control. Consider, for example, if you are reading the new operations manual used at work and notice that many procedures have changed. Perhaps you realize that you may not remember these changes when you need them (self-monitoring). To prevent a memory failure, you decide to create a list of keywords for each procedure (self-control) and to keep it in your pocket (or paper or electronic planner) during work hours.

Thus, strategy decisions are linked to self-monitoring for tasks that are complex and non-routine. For repetitive everyday tasks, strategy decisions tend to be routine, without much conscious control. For a more complete description of the relationships between beliefs, self-monitoring, and self-control, see Kennedy and Coelho (2005).

Assessing Memory and Metamemory

SLPs are a part of an interdisciplinary team that assesses memory skills. The assessment process typically includes formal, standardized evaluation tools that provide information about the individual's memory impairment. For practice guidelines on the use of standardized assessment for individuals with TBI, readers are referred to Turkstra et al. (2005).

Informal assessment provides SLPs with information about individuals' awareness of their memory problems and whether everyday forgetting is related to poor self-monitoring or self-control. Self-other questionnaires, structured interviews and systematic observation during functional activities are particularly useful for assessing TBI survivors' beliefs about their memory. For example, the Everyday Memory Questionnaire (Sunderland, Harris, & Baddeley, 1983) can be filled out by the TBI survivor and a family member or clinician independently and then compared to see if their perceptions match. (Clinicians should be aware that not all individuals are good judges of others' behavior. This is a caveat to the use of self-other questionnaires for assessing awareness.)

Interviews can be conducted by gradually asking more specific questions about clients' perceptions of their memory. Clinicians can begin with a question such as, "Do you have any memory problems?" and gradually asking such questions as, "What types of things do you forget? Do you use anything to help you remember and if so, what are they?" These queries provide SLPs with information about the breadth and depth of a client's awareness. By adding a Likert rating scale to the interview, valuable information can be obtained about how certain a client feels about the answers. Clinical observation of what the client does during complex activities that tax memory then should be compared to questionnaire information and answers from the interview.

Treatment Principles and Techniques

The mechanic knows that having the right tools for the job is important. The mechanic also knows that certain tools are better suited than others for specific tasks. Having tools, however, does not mean you will use them.

• External Memory Aids. Like the mechanic, individuals with TBI need a toolbox of strategies to help them avoid various kinds of memory failure. External memory aids are "supports" used to compensate for forgetfulness. These aids include everything from signs, calendars and daily planners (including PDAs), alerting systems, procedural checklists, etc. External memory aids are effective in preventing memory failure when a personalized approach is used in which clients help select the aid, while the clinician shapes its use and fades the cues. For a description of evidence for the use of external memory aids, readers are referred to Sohlberg et al. (in press).
• Internal Memory Aids. Internal memory aids are mental strategies used to enhance memory and improve the likelihood that information will be recalled later. These include visual imagery, elaborative encoding, grouping, etc. Self-created personalized visual images can be effective in improving recall for specific kinds of activities, such as remembering someone's name (e.g., Kashel et al, 2002). For example, if trying to remember the name "Cliff," an internal image of a person standing at a cliff's edge could facilitate memory. Elaborative encoding involves associating elements or features with the to-be-remembered item (e.g., Oberg & Turkstra, 1998). Grouping or chunking can be used when a lot of material needs to be remembered. By organizing the information into logical categories, individuals are more likely to remember the category, which then facilitates recall of the items within that category.

Metamemory Intervention

Impaired metamemory will affect whether or not memory aids are used. Alternatively, knowing that you have memory problems does not mean you will automatically use the aids introduced in treatment. For a description of evidence for intervention for executive function and metacognitive disorders, readers are referred to Kennedy et al (under review). Several intervention techniques are available that can facilitate accurate metamemory in survivors of TBI and thereby improve the likelihood that memory aids will be used:  

• Individualize educational information about the client's specific memory and cognitive disabilities. There is little evidence that self-awareness improves when an individual is presented with general educational information. Individuals with TBI need individualized, specific information about their abilities and disabilities, so that they do not have to figure out what applies to them and what does not.
• Involve the client in setting memory goals and in selecting memory aids. See Webb and Gluecauf (1994) for an example of high-goal involvement and low-goal involvement. Those who were highly involved were better at setting goals two months after treatment ended, compared with those who were less involved.
• Create opportunities for accurate self-monitoring. Regardless of brain injury or frontal lobe injury, adults are very accurate at predicting their memory when the predictions are slightly delayed (Kennedy & Yorkston, 2004). When predictions are made immediately after or during studying (within about 30 seconds), our predictions are barely above chance. Clinicians can create an opportunity for accurate self-monitoring by forcing a delay between, for example, studying a grocery list and asking clients to predict how many items they will remember. If they predict that they will remember very few items, they can be directed by the clinician to use a memory aid. Thus, the link between self-monitoring and self-control is made explicit. Additionally, by showing the client how accurate they can be when making delayed predictions compared with immediate predictions, we give the client positive, direct feedback that they can be "good" at judging their learning. See Dunlosky, Hertzog, Kennedy & Thiede (2005) [posted online] for a description of a self-monitoring approach to effective learning across adult populations.
• Integrate metacognitive strategies into training individuals to use memory aids. Make explicit the link between self-monitoring (e.g., predictions) and strategy decisions. Modifying instructional sequences to individuals' needs will help to upgrade beliefs about memory and will improve memory. One example of such a sequence is as follows: Skim the material, make memory predictions and create a study plan, carry out the study plan, take a self-quiz, compare results with predictions, update predictions and plan. See Kennedy, Carney, & Peters (2003) for an example of the memory benefits to adults with brain injury when study strategies are linked to accurate self-predictions.
• Provide distributed practice at high levels of accuracy. Practice with strategy supports in a sequence of steps is critical for individuals with memory impairment. Building in breaks in between practice sessions (i.e., distributed practice) increases the likelihood that the person will use the memory strategy or recall the information. Shaping correct use of supports through cues or prompts will foster high levels of accuracy. Fading the cues while maintaining accuracy is demonstrated in a technique called "spaced retrieval." See Sohlberg, Ehlhardt, & Kennedy (2005) for a discussion of instructional techniques that are built on errorless learning and spaced retrieval.

As with other cognitive-communication disorders, individualizing treatment approaches is critical to clients' success. Actively involving our TBI clients in treatment strategy decisions will provide them with the metacognitive tools to tackle situations that will arise in the future when the SLP is not present to assist the client.