This week, I have chosen to expand the working model of memory (Baddeley & Hitch, 1974), to include the phenomenon semantic satiation (Smith & Klein, 1990). The working model of memory (Baddeley & Hitch, 1974) has become somewhat of a stalwart at the heart of extant literature on human memory, after having succeeded the multi-store model of memory put forward by Atkinson and Shiffrin (1968). The multi-store model fell into disfavour because its accounts of short-term and long-term memory were over-simplified (Eysenck & Keane, 1995). Baddeley and Hitch (1974; Baddeley 2000) argued that the concept of short-term store should be replaced with that of working memory. In their professional judgement, the working memory system consisted of four components:
- A modality-free central executive resembling attention.
- An articulatory or phonological loop which holds information in a phonological form.
- A visuo-spatial sketch-pad which is specialised for spatial and/or visual coding.
- An episodic buffer, where information from different modalities and sources are bound together to form new episodic memories.
Eysenck & Keane (1995)
The present article is most interested in the articulatory or phonological loop component of working memory. Elaborating upon the description given above, Ashcraft and Radvansky (2010) describe it as “the speech- and sound-related component responsible for rehearsal of verbal information and phonological processing”. The phonological loop can be divided further into another two components, the phonological store and the articulatory loop. The phonological store, sometimes referred to as the ‘inner ear’ is essentially a passive store component of the phonological loop; this is the part that holds onto verbal information (Ashcraft & Radvansky, 2010). However, information in the phonological store will be forgotten unless it is actively rehearsed and refreshed. Thus, the rehearsal is the role of the articulatory loop, or the ‘inner voice’, the part of the phonological loop involved in the active refreshing of information in the phonological store (Ashcraft & Radvansky, 2010). A number of effects pertinent to the workings of the phonological loop have been established, such as the word length effect, the articulatory suppression effect and the phonological similarity effect (Ashcraft & Radvansky, 2010). However, the purpose of this article is not to go into depth explaining these effects, but to provide insight into another effect pertinent to the functioning of the phonological loop, that has seemingly been neglected by current curriculum and textbooks in the field.
One of the principal tenets of the working memory model is that, repetition or rehearsal enhances memory. In fact, this tenet had consonantly been adopted into the doctrines relating to the functioning of human memory, long before the emergence of the working memory model (Ebbinghaus, 1913; Crowder 1976). However, it is important to consider that there is a small but robust body of evidence suggesting that this is not always the case. That is, mass repetition can actually reverse the beneficial effects of brief repetition.
This effect, or phenomenon has persisted to torment me throughout my undergraduate study, it has been termed semantic satiation and described as “the subjective experience of loss of meaning of a word as a result of prolonged inspection and repetition of that word” (Smith & Klein, 1990).
It is common for me whilst revising for an upcoming exam, to compile lists of information and spend substantial amount of time rehearsing these lists. Rationale for such list rehearsal can be attributed to last year’s cognitive psychology module. It can, more specifically, be traced to an assignment where we were required to compile a list of revision tips. One of the tips included on my list was ‘rehearsal’, referring to the beneficial effect of rehearsal on the encoding of to-be-learned information. However, this ‘tip’ has caused me some difficulty. Adhering to the tenet ‘the more rehearsal the better’, I found that rehearsing lists of information intensively actually had an adverse effect, culminating in an eventual temporary loss of meaning for the to-be-learned information. This experience was largely unsettling for me and a phenomenon that I thought was exclusively effecting me. It was not until I incidentally stumbled across literature examining semantic satiation that I realised I was not alone in experiencing such difficulties.
Prior to research by conducted by Smith (1984), severe methodological issues hindered the progression of literature examining semantic satiation. This may account for why semantic satiation remains largely an alien concept to many.
Smith (1984) developed a paradigm in which participants repeated semantic category names (e.g. tools) for either 3 or 30 consecutive repetitions. Participants were then instantaneously presented with a target that was either a member (e.g. hammer) or a non-member (e.g. banana) of the category, and then required to make a category membership decision immediately after. The key finding of the study was that category membership decisions were actually slower following 30 repetitions than after only three.
It is important to note at this stage of the article, that extant literature has typically viewed the effects related to semantic satiation to be very brief or short-lived; thus, neglecting to examine whether semantic satiation has prolonged effects on long-term semantic accessibility. Resultantly, the durability of semantic satiation remains unclear. If such a persisting effect could be shown, it would constitute a striking exception to the general principle that repetition improves long-term memory for repeated materials (Kuhl & Anderson, 2011). It is rarely considered, that prolonged rehearsal time may actually reverse the benefits to long-term memory that are typically associated with brief rehearsal.
In their article, Kuhl and Anderson (2011) highlight a case in which repetition is not better. The case of interest is that of massed rote rehearsal and its relationship with long-term semantic accessibility. To test this, Kuhl and Anderson (2011) adopted a paradigm in which participants were first required to repeat words (e.g. sheep) one at a time for intervals of 0, 5, 10, 20, or 40 seconds. Followed then by a free-association phase where participants were provided cues that could be completed for repeated words (e.g. “herd s___” for “sheep”) or with semantically associated words (e.g. “fabric w___” for “wool”). Brief periods of repetition (5-10 seconds) resulted in priming, as would be expected based on research on repetition priming and spreading activation. Longer periods of repetition (20-40 seconds), however, abolished priming. Interestingly, this massed-repetition decrement was particularly robust for semantic associates of repeated words, and was evident after a 15 minute delay. When ideas are repeated often enough, the benefits of rehearsal can actually be reversed.
The finding that massed-repetition decrement was evident after a 15 minute delay indicates that the effects of semantic satiation are durable, having prolonged effects on the accessibility of semantic information from the long-term memory. Furthermore, this finding could provide evidence refuting the theory that ‘rehearsal’ improves long-term memory for repeated materials.