Introduction
Relevant cues are expected to improve retrieval from memory. They are assumed to jog weaker memory traces that cannot be activated on their own and make them accessible to the rememberer (e.g., Tulving & Pearlstone, 1966). Indeed, the expectation of beneficial effects of cuing is so ingrained that when cues do not facilitate remembering, it is considered to be a memory quirk. In some cases, when these findings are explored further rather than being ignored as glitches, however, our knowledge of memory can be fortuitously advanced.
For instance, one such quirky finding was reported in the late 1960s and opened up a brand-new avenue for investigating memory inhibition. When some of the items from a studied list were given as cues to remind the participants of the rest of the items on the list, recall of the rest of the items was not facilitated as expected but actually inhibited (Slamecka, 1968, 1969). A similar effect emerged in semantic memory, as well, when US states were the to-be-recalled items (Brown, 1968). These surprising findings, now known as the part-set cuing effect, were subsequently replicated many times and led to a variety of theoretical explanations and valuable empirical findings (e.g., Nickerson, 1984; see also Chapter 8 this volume). In fact, the effect was even shown to be a phenomenon that went beyond memory and touched cognition in general (PeynircioÄlu, 1987a). Today, the part-set cuing effect has a prominent place in memory as a well-established quirky phenomenon, which, even if not fully explained yet, is still being investigated and tweaked (e.g., AndrĂ©s & Howard, 2011; Fritz & Morris, 2015) while also serving as a springboard for explanations of other findings, such as the collaborative inhibition effect (e.g., Basden, Basden, Bryner, & Thomas, 1997).
Cues for jogging memory can be extrinsic to the to-be-recalled items or contextual in nature. For instance, providing a category name (e.g., Pollio & Gerow, 1968; Tulving & Pearlstone, 1966), reinstating the context as in state-dependent learning studies (e.g., Eich, 1980), or providing another item related to the target, semantically (e.g., Hudson & Davis, 1972) or otherwise, such as the preceding item or a paired-associate item (e.g., Ebbinghaus, 1885; Tulving & Thomson, 1973) helps jog memory for the to-be-recalled item. These findings are easily explained through the widely accepted view that memory is associative. Cues can also be intrinsic to the to-be-recalled items or item-specific, often exemplified by fragment cuing or stem completion procedures, in which some of the letters of a word are given as cues (e.g., Nelson, Canas, Bajo, & Keelan, 1987) or pictures are presented with missing parts (e.g., Luo & Snodgras, 1994). In the case of words, at least in the absence of extrinsic cues (cf. Tulving & Thomson, 1973), recognition tests are assumed to provide the ultimate intrinsic cues because, with all the letters present, items themselves are the most complete cues that could possibly be presented. This distinction between extrinsic and intrinsic cues has also been discussed using the labels ârelationalâ and âitem-specific or item-basedâ information and incorporated into theories of remembering (e.g., Gillund & Shiffrin, 1984; Guynn et al., 2014). Although extrinsic or contextual cues have tended to capture the lionâs share of both theoretical and research efforts through the years (e.g., Bower, 1970; McGeoch, 1942; Raaijmakers & Shiffrin, 1981; Tulving & Osler, 1968), in both cases, regardless of whether the cues are extrinsic (or contextual/relational) or intrinsic (or item-specific) to the to-be-recalled items, their function is to prompt memory (e.g., Mulligan & Peterson, 2015; Peterson & Mulligan, 2013).
Cue depreciation, the topic reviewed in this chapter, involves a quirky finding in the verbal memory domain, which involves the latter type of cues, those that are item-specific or intrinsic to the to-be-recalled items. In particular, it involves presenting various letters of the words as cues. When these fragments are given incrementally or in an ascending fashion, they are less effective in prompting memory than when they are given all at once (e.g., PeynircioÄlu & Watkins, 1986), but only in certain situations such as when the words have been studied or primed in some way. For instance, if a studied word such as RAINDROP is cued by first presenting the fragment R- â - â - -P and then gradually increasing the number of letter cues to R â I â - â - P then to R â I â - R â P and then to R â I â - ROP, participants are less likely to discover the word RAINDROP compared to when it is cued using only the last fragment (R â I â - ROP) presented for the same amount of time as any of the other fragments in the incremental case. Note that the last fragment of the ascending series is exactly the same as that in the single-cue or standard condition and is presented for exactly the same amount of time, hence the designation as quirky. Thus, one might expect the preceding fragments in the incremental condition, which are simply subsets of the larger fragment, to provide even more time to think about the target wordâand it is important to note that performance is calculated cumulatively in the incremental condition (that is, any recall prompted by the lesser cues is added to that prompted by the last fragment). But instead it appears that more cues lead to less recall.
Despite intermittent research on the cue depreciation effect a fully satisfactory explanation still eludes us after more than 30 years since it was first reported. This chapter briefly reviews the phenomenon, highlights its quirky aspects, and speculates about further studies that could shed more light on the underlying mechanisms. In the following sections, first this quirky effect is differentiated from two other effects that involve situations when more cues lead to less recall. Then, the initial findings and implications of the cue depreciation effect are presented followed by studies on some boundary conditions and the generality of the effect. Findings from these studies help delineate the effect in more detail as well as guide the emergence of various explanations, which are then reviewed in the next section. Finally, the last section discusses the current state of the quirky effect and offers some conjectures about what it might all mean as well as some ideas for future research.
...