Phonological Neighborhood Effect on Short-Term Memory for Order

The company a word keeps: The role of neighbourhood density in verbal short-term memory

“…These trends have been suggested to reflect easier item encoding for visually presented dense neighbourhood words, which in turn leads to better memory for those items (e.g. Clarkson et al, 2017 suggest that more resources may be left available for order encoding). If these trends are indicative of encoding costs/benefits for dense and sparse neighbourhood words in the different modalities then the consequence should be an attenuated, or perhaps even reversed, NDE when using auditory presentation.…”

“…There was no significant difference between the pools on word frequency, t(94) = 1.25, p = .21, or between the words which have imageability values associated with them, t(62) = 0.06, p = .95. 3 However, of importance to note here are that the materials used by Clarkson et al (2017) were originally controlled for phonological but not orthographic ND. As noted earlier, phonological and orthographic neighbourhood densities are highly correlated (Mulatti et al, 2003) and, as such, the current pools vary on both dimensions.…”

“…To-be-remembered lists comprised of words with many lexical neighbours (i.e. from a dense neighbourhood) are better recalled than lists comprising words with fewer lexical neighbours (Allen & Hulme, 2006;Clarkson et al, 2017;Derraugh et al, 2017;Guitard, Gabel, et al, 2018;Jalbert, Neath, Bireta, & Surprenant, 2011;Roodenrys et al, 2002;Vitevich et al, 2012).…”

“…Alternatively, ND supports encoding rather than retrieval. Clarkson et al (2017) described two processes through which ND could act in vSTM. One relies upon order encoding (DeLosh & McDaniel, 1996) and supposes that item and order information are encoded separately but act jointly to bring about retrieval of an item.…”

“…This increased activation promotes the easier encoding of item information, liberating more resource for the encoding of serial order of words from dense neighbourhoods, thereby reducing the likelihood of errors at recall. The second explanation offered by Clarkson et al (2017) is that during encoding LTM supports item-to-position associations increasing correct serial position recall. Because dense neighbourhood words elicit more activation within LTM their item-to-position associations will be stronger than those for sparse neighbourhood words.…”