What is the process of encoding memory?

2.07.3.1 Intent to Remember

Intuitively, intent to remember emerges as a dominant factor affecting later memory, but research on encoding processes has shown that intuition unequivocally to be wrong (e.g., Postman, 1964; Hyde and Jenkins, 1969; Craik and Lockhart, 1972; Challis et al., 1996). As we shall see, what does matter is the type of processing performed on the material, but trying to remember does not ensure that optimal processing will be engaged. Figure 3 depicts the results reported by Hyde and Jenkins (1969). Subjects were asked to determine the words’ pleasantness, check all of the ‘e’s in the word, or count the vowels in each word. For three groups of subjects, the orienting tasks were given as incidental memory instructions, and for another three groups the orienting tasks were accompanied by instructions to try to remember the words. As can be seen in Figure 3, adding intentional instructions improved performance in the nonsemantic orienting groups but had no effect on the performance following the pleasantness rating task. It is not the intent to remember but, rather, the nature of the processing that is important.

Indeed, an enduring contribution of levels of processing (Craik and Lockhart, 1972) is the acceptance of memory as a by-product of the processes of perception and comprehension of the original experience rather than as the intentional object of processing. After all, how many times during the course of the day does one try to remember, and yet healthy adults can remember most everything that happened yesterday. Furthermore, only occasionally do we know what, if anything, about current experience will be required from memory, rendering intent to remember any part of the experience a gamble against future demands. In light of these considerations, the lack of direct effects of intentional memory is understandable.

2.10.1.4 Retrieval Cues

As described earlier, organization, elaboration, and mental imagery are often treated as part of the encoding process. However, one could just as easily suggest that these processes are part of the retrieval process. Indeed, as Tulving (1983) convincingly argued, any distinction between encoding and retrieval processes is purely heuristic. Processes occurring at the time of learning exert their effect on the retrieval process, either as facilitation or interference, and thus no meaningful theoretical dichotomy between encoding and retrieval remains coherent. The challenge is to explain the influence of processing at the time of initial experience on processes required for successful test performance. To this point, we have hinted at the importance of developing cues that are diagnostic of the to-be-remembered information. A more elaborate rendition of this idea is that the processing of the original experience, when reinstated at testing, constrains production to a limited set of items. This idea has been suggested to explain the effect of various encoding manipulations.

An example of this type of approach within the context of mnemonics is research (i.e., Wallace and Rubin, 1991; Rubin, 1995) that has examined the use of rhyme and meaning to cue memory for narratives in the oral tradition. Such research suggests that rhyme and meaning cues work in concert to facilitate memory by constraining the number of stored choices available at retrieval (cf. Rubin and Wallace, 1989). More generally, this notion of constraining choices at retrieval is at the heart of the effectiveness of distinctive processing. As noted by Hunt and Smith (1996), organization serves to specify the episodic context in which to-be-remembered information was embedded. The addition of item-specific information (e.g., a unique cue) along with organizational processing limits the retrieval set to items that both share the unique feature and were present in the specified context. Thus, processes that ostensibly occur during encoding may very well exert their influence by providing diagnostically precise cues at retrieval.

1 Different Kinds of Reasoning Processes

According to Sternberg (1986), reasoning involves component processes that serve three types of functions. Selective encoding processes distinguish relevant from irrelevant information in the stimulus and store the selected information in working memory. Selective comparison processes are responsible for determining which information from long-term memory is relevant for solution and will be retrieved and stored in working memory. Selective combination processes analyze, manipulate, and integrate the information placed in working memory by selective encoding and comparison processes.

At this level no commitment is made about what the component processes are that constitute selective encoding, comparison, and combination, or how these components are organized; that is, these three types of functions are computational requirements rather than algorithmic specifications. In Sternberg's (1986) theory, selective encoding, selective comparison, and selective combination are best viewed as functions that are served by component processes that may vary in type and organization from task to task.

B Stimulus-Specific Organization

Stimulus-specific organization refers to the structure of individual visual stimuli. Research to date has focused on the effect on encoding processes of varying the structure of such stimuli. The first to investigate the phenomenon in persons with mental retardation were Caruso and Detterman (1983). The stimuli used by these investigators were 16-cell checkerboard-type matrix patterns (Fig. 3). These stimuli are particularly informative in that three structural variables may be manipulated within each stimulus; symmetry, number of adjacencies, and number of cells filled. Caruso and Detterman (1983) concluded from their study that persons with mental retardation and their nonretarded, CA-matched peers processed the stimuli in a similar manner; however, inspection times were unlimited, thus minimizing memory demands, and a four-choice match-to-sample format was used that did not allow for direct examination of target–distractor disparity effects.

Using similar checkerboard-type stimuli, Soraci, Carlin, Deckner, and Baumeister (1990) assessed the relation between stimulus organization and encoding in persons with and without mental retardation, with CAs between 15 and 18. Three important methodological differences existed between this study and that of Caruso and Detterman (1983). First, stimulus organization was manipulated across high, moderate, and low levels of structure. This allowed for a direct comparison of the encodability of forms varying in structure for each of the groups. Second, a two-choice match-to-sample format was used to permit direct study of target-distractor disparity effects. Finally, memory demands were increased by limiting target stimulus exposures to less than 150 milliseconds and instituting a 1-second interval between sample offset and presentation of the two comparison stimuli.

Results demonstrated that persons with and without mental retardation more effectively detected stimuli when there was either high structural organization (e.g., symmetry) of the target or distractor stimulus or increased target-distractor structural disparity (e.g., highly structured target stimulus and low-structure distractor stimulus). Increased disparity has also been shown to enhance discrimination learning (Dinsmoor, 1985) and proficiency of visual search (Spitz & Borland, 1971) in persons with and without mental retardation.

An important intelligence-related difference was identified. Under conditions of reduced interstimulus disparity (e.g., with a moderate-structure distractor), subjects with mental retardation performed much more poorly than did the nonretarded subjects (Fig. 4). The individuals with mental retardation thus exhibited a lower sensitivity to the available relational information. This information was specified by conjunctions of symmetry and number of adjacencies within the individual stimuli. These findings suggest that the differential detection of moderate levels of stimulus organization could be responsible for intelligence-related differences on a wide range of tasks, as most experimental and “real-world” stimuli are composed of mixed organizational parameters (Soraci, Carlin, Deckner, & Baumeister, 1990).

A further study (Carlin & Soraci, 1991) was designed to determine whether similar intelligence-related differences would be exhibited with formlike stimuli varying solely with respect to symmetry; that is, stimuli were either vertically symmetric, horizontally symmetric, doubly symmetric (i.e., both vertical and horizontal), or asymmetric (Fig. 5). Methods used were identical to those in the Soraci, Carlin, Deckner, & Baumeister (1990) study. Vertically symmetric stimuli were processed more efficiently than the other types of symmetric stimuli. Interestingly, no intelligence-related differences were found. The lack of intelligence-related differences indicated that, at least for manipulations involving symmetry, the groups were equally sensitive to the relevant dimension of difference and the relational information embedded in the task. These results are at odds with previous findings (Soraci, Deckner, Baumeister, & Carlin, 1990) that indicated a lower sensitivity to relational information in persons with mental retardation on an encoding task. This discrepancy may have resulted from the fact that the perception and discrimination of symmetries per se, in the absence of internal structure (i.e., contiguity), are more basic processing abilities.

The results of these studies indicate that persons with mental retardation are able to perform fairly well on a speeded encoding task when stimuli vary along only a single dimension (e.g., symmetry). When stimuli are defined by a conjunction of symmetry and number of adjacencies, however, subjects with mental retardation perform more poorly than CA-matched peers under conditions of reduced interstimulus disparity. This group difference makes salient the locus of an encoding deficiency, which involve central scanning mechanisms, in persons with mental retardation.

2.1 Pharmacological Modulation of 5-HTRs During the Encoding Phase

Among 5-HTRs, the subtypes 5-HT1A, 5-HT4 and 5-HT6 are the three receptors for which involvement in the encoding process has been reported. Concerning 5-HT1AR, its blockade prior to the sample phase facilitated encoding. Indeed, using a condition that favoured rats spontaneous forgetting (intertrial interval of 24 h), Pitsikas and colleagues demonstrated that presample administration of WAY-100635 (5-HT1AR antagonist, 1 mg/kg; i.p.; 30min prior to the sample phase) increased the discrimination index (Pitsikas et al., 2003). Afterwards, similar results were confirmed by another research group and were associated to a potentiation of the learning-induced increase in AMPA receptor subunits (Schiapparelli et al., 2006). Conversely, for 5-HT4R, the main beneficial effect was reported after the activation of this receptor. RS-67333, a widely used 5-HT4R partial agonist, administered 30 min before the acquisition phase at the dose of 1 mg/kg, i.p., improved object recognition at 24 h ITI (Lamirault and Simon, 2001). Of note, the concomitant administration of a selective 5-HT4R antagonist with the agonist RS-67333 fully abolished the memory improving effect of the agonist alone, which confirms the involvement of 5-HT4R (Lamirault and Simon, 2001; Moser et al., 2002). Besides using a more recently designed 5-HT4R partial agonist (SL65.0155), similar cognitive enhancing effect was again found in the rat (Moser et al., 2002). However, it must be mentioned that, in this last study, pharmacological administrations were performed before each of the three experimental phases of the test (habituation, sample phase and choice phase). Such a procedure does not allow to address which process is affected by the treatment. Concerning the 5-HT6R, although they are somehow controversial, data from literature argue for a cognitive enhancing effect when the receptor is blocked during the pre-sample phase (Gravius et al., 2011; King et al., 2004; Lieben et al., 2005; Thur et al., 2014). Indeed, while all investigations were performed in the conditions that enable spontaneous forgetting in rats, several intertrial interval (ITI) were used, ranging from 4 to 24 h. Thus, presample administration of 5-HT6R antagonists, Ro 04-6790 or SB-271046 (10 mg/kg i.p. for both drugs; 20 min prior to sample phase) prevented the spontaneous forgetting observed with a 4 h ITI (King et al., 2004). Two other 5-HT6R antagonists (Ro 4368554 and SB-258585; 1–3–10 mg/kg i.p.; 60 min prior to sample phase for Ro 4368554; 3–10–30 mg/kg; i.p.; 90 min before sample phase for SB-258585) also improved recognition memory with a 6 h ITI (Gravius et al., 2011). However, with a 24 h ITI, Ro 4368554 (1–3–10 mg/kg; i.p.; 60 min prior to sample phase) failed to induce cognitive enhancing effect (Lieben et al., 2005). The most controversial result comes from the investigation of Thur and colleagues (Thur et al., 2014), who did not find a memory enhancing effect of Ro 04-6790 (5–10 mg/kg i.p.;, 20 min prior to sample phase) with a 4 h ITI. While it has been suggested that differences in experimental designs could account for such discrepancies (such as strain or age of rats), we would like to point out that, in this last study, a possible sedative effect of Ro 04-6790 could have prevented a beneficial cognitive effect on recognition memory. However, paradoxical results were reported in a study by Kendall et al. (2011). The stimulation of 5-HT6R by the selective agonists E-6801 and EMD-386088 induced the same procognitive effects than that obtained with antagonists. Indeed, with a presample administration, the authors have demonstrated that E-6801 (from 2.5 to 10 mg/kg i.p.) and EMD-386088 (5 mg/kg i.p.), two 5-HT6R agonists, were able to counteract delay-dependent object recognition discrimination deficit (ITI 4 h) in rats. In this study, the same result was obtained with the two selective 5-HT6R antagonists SB-271046 (10 mg/kg i.p.) and Ro 04-6790 (5 and 10 mg/kg i.p.). The authors speculated that the 5-HT6R agonists and antagonists could both induce an increase of glutamate and/or acetylcholine through different mechanisms (agonist could activate 5-HT6R located on glutamatergic and/or cholinergic neurons while antagonists could block 5-HT6R located on GABAergic neurons disinhibiting glutamatergic and/or cholinergic neurons). Such an increase in glutamate and/or acetylcholine could then improve the cognitive performances.

It is worth noting, however, that cognitive enhancing effects have been reported after blockade of 5-HT3R (Bétry et al., 2015). Indeed, presample administration of vortioxetine (a multimodal antidepressant that acts as a mixed 5-HT3R, 5-HT7R and 5-HT1DR antagonist, as a 5-HT1BR partial agonist as well as an inhibitor of the 5-HT transporter) improved object recognition in rat (ITI 24 h). Involvement of 5-HT3R was confirmed by presample acute administration of SR-57227 (a 5-HT3R agonist), which prevented cognitive enhancement by vortioxetine, showing that 5-HT3R is involved in the benefits of vortioxetine on recognition performance.

1 Firing Pattern

An effective way for a drug of abuse like alcohol to modify temporarily or permanently messages encoded by neurons is to act on the ultimate form of the neuronal encoding process, i.e., the action potential. BK channels, among other ion channels underlying action potentials, warrant particular attention as they are activated within a range of potentials (–60 to +40 mV) that correspond to the ascending phase of the action potential. The influx of calcium through voltage-gated calcium channels (VGCCs), activated by rapid depolarization induced by fast Na+ currents, increases the open probability of these channels. BK channel activation in turn helps the cell repolarize and deactivates VGCCs, creating a powerful feedback mechanism controlling action potential duration and firing patterns (Brenner et al., 2005; Edgerton and Reinhart, 2003; Faber and Sah, 2002, 2003). An elegant single-channel study carried out in cell-attached and inside-out patch-clamp recording modes in hippocampal neurons (Marrion and Tavalin, 1998) demonstrated that BK channels and voltage-gated N-type calcium channels co-localize on the same patch of membrane and that calcium influx through these calcium channels provides the source for BK activation, confirming a symbiotic relationship between these two types of ion channels (see also Loane et al., 2007). In addition to modulating action potentials in the brain (Robitaille and Charlton, 1992; Shao et al., 1999; Storm, 1990; Widmer et al., 1998) and peripheral nervous system (Gruss et al., 2001; Scholz et al., 1998), BK channels also contribute to action potential after-hyperpolarization (AHP) (Adams et al., 1982; Goldberg and Wilson, 2005; Lancaster and Nicoll, 1987; Shao et al., 1999). Interestingly, the modulation of action potentials by BK channels does not appear to be limited to somatic action potentials. Evidence from immunohistochemical and electron microscopy studies suggests that BK channels are expressed in the membrane of axons and terminals (Kaufmann et al., 2009; Knaus et al., 1996; Misonou et al., 2006; Sailer et al., 2006) where they control action potential repolarization (Dopico et al., 1996).

2 Internal Strategies

Internal memory strategies include techniques to improve encoding and retrieval. Like other skills, these strategies must be practiced over time. Furthermore, to execute such skills, one must be able to focus attention on the task at hand.

Many internal strategies target the encoding process. Through various methods, associations between new information and old (i.e., semantic) knowledge are formed in ways that make sense to that individual. These new associations provide cues for later retrieval. Because these strategies depend upon previously stored semantic knowledge, in cases where semantic memory is impaired, internal strategies might not be effective. Three common internal strategies are imagery, verbal elaboration, and organization (West 1995). As noted earlier, internal strategies are often not successful for people with significant consolidation problems, as they will not remember to employ them when necessary (Wilson and Moffat 1984, Glisky 1995).

Imagery involves the association of interactive visual images. This method has proven especially effective in the learning of names and faces (Yesavage et al. 1983). In this instance, one would study a new person to identify a characteristic that could be linked to that person's name. For example, when meeting a person named ‘Karen Singer,’ one could conjure up an image of the person singing. This technique can also be used to link words together, such as items from a grocery list, errands that must be done, or daily chores. For example, if one needs to pay the bills and do the vacuuming, one could create a visual image of vacuuming away the bills. A single image would therefore provide the cues for remembering both chores.

Verbal elaboration involves the creation of new words or sentences, rather than images, to link information together. These techniques are dependent upon semantic and verbal abilities. If one wanted to remember a new name, one could combine the first and last name into a new sentence. For example, the name ‘Darren Wernick’ could be remembered with the sentence ‘Darren wore a neck-tie,’ as the words ‘wore’ and ‘neck’ would provide the cue for the name. Mentally linking this sentence to a visual image, such as Darren putting on a tie, would further strengthen the elaboration and enhance later retrieval. Making up rhymes is a common method for recalling information (e.g., Thirty days hath September …). Another method is first-letter elaboration, in which one uses the first letters in words to create a new word or pseudo-word. For example, to remember emergency procedure for a hospital fire, employees are taught to remember the word ‘RACE.’ This not only reminds workers of the steps to take in case of a fire, but also the order of steps (‘Rescue’ patients, sound the ‘Alarm,’ ‘Confine’ the fire, ‘Evacuate’ the premises).

It is important to note in both imagery and verbal elaboration that the imaginary links created between objects or words are most effectively remembered when they make personal sense to that individual, even if they seem nonsensical to others. Often the first and most creative idea is the most salient cue for later recall.

Finally, organization of incoming information is beneficial to emphasize natural links between stimuli in order to enhance later retrieval. Organizational techniques can be particularly useful in situations where the individual has to learn information that exceeds attention span limitations (e.g., a lengthy list). One effective method of organization is ‘chunking,’ or grouping items into categories (Cermak 1975). When trying to remember items for the grocery store one could organize the list so that the five dairy items and the five vegetables are grouped together. Therefore, simply remembering the two general categories of dairy items and vegetables would enhance memory retrieval. This technique is also used to remember sequences of numbers. For example to remember a phone number, one could chunk ten individual numbers into three larger numbers.

5.1 Early Recognition and Intervention

In looking for signs of the onset of dementia, it is important to differentiate between normal changes in memory that occur with aging and changes that may be associated with dementia such as Alzheimer’s disease. In normal aging, the older person takes longer to learn new information, and the encoding process requires more time for new material to be retained, than is the case in younger persons. Therefore, an apparent inability to learn or to remember what was just heard is not necessarily a sign of early dementia; rather, it may be a sign of slowed cognition that occurs in all aged persons.

Changes with dementia, on the other hand, emerge with difficulty in learning new information. The person often gives up trying to learn. Difficulty in learning subsequently affects the ability to recall, that is, memory. This applies to verbal information as well as spatial information such as remembering directions.

1.1 The Trace-dependent View of Retrieval

The simplest account of memory retrieval assumes that it depends only on the strength of the memory trace relative to other traces. Retrieval failure occurs because the memory trace has become too weak, or because competing traces have become stronger than the target trace. Such a trace-dependent view of retrieval held sway in much of experimental psychology until well into the 1960s.

According to the trace-dependent view the strength of the memory trace (and thus the likelihood of retrieval) is a function of its initial strength and the length of time between this original experience and the attempted retrieval. The body of evidence supporting both these claims is overwhelming. The second assumption is supported by countless experiments that show a gradual decrease in retrieval success as the retention interval increases. Consider now the first assumption. The initial strength of a trace is determined by a number of factors, most notably by the way in which the experience is encoded, and it is not at all difficult to demonstrate that such encoding processes influence retrieval (see Memory: Levels of Processing). For example, in experiments requiring the recall of names of simple objects, recall levels can be greatly enhanced if, during presentation of the names, participants are asked to form a visual image of each object.

The trace-dependent view can account for many of the basic phenomena of memory retrieval. For example, the fact that retrieval becomes more difficult with passing time is a consequence of a weakening of trace strength. The fact that an unrecallable item can nevertheless often be recognized is explained by the claim that recall requires a stronger trace than does recognition. By appealing to fluctuations in trace strength over time it can even account for the common experience of being unable to recall something at one moment but being able to do so at a later time.

What is the encoding process?

What are the four processes of encoding?

What are the processes of memory storage?

What are the two major processes of encoding?