|
Article Excerpt
Evidence from several sources shows that adults find it harder to remember names than other types of biographical knowledge about familiar people. This has been demonstrated in natural settings, where people frequently report experiencing difficulties in retrieving someone's name whilst being able to remember other information such as the person's occupation or where they would usually be seen (Cohen & Faulkner, 1986; Young, Hay, & Ellis, 1985). The effect is also observed in laboratory studies where it is common for participants to be able to recall someone's occupation while being unable to retrieve their name (e.g. Hanley & Cowell, 1988; Hay, Young, & Ellis, 1991).
Further evidence for the difficulty of name retrieval comes from reaction time studies showing that face naming takes longer than categorizing faces by their occupation or nationality (e.g. Young, McWeeny, Ellis, & Hay, 1986). Even when response requirements are equated across tasks, difficulties in name retrieval still persist. Matching tasks have shown that it takes longer to decide whether two faces share the same first name than it does to decide if they share other characteristics such as occupation, nationality or being alive/dead (Carson, Burton, & Bruce, 2000; Johnston & Bruce, 1990; Young, Ellis, & Flude, 1988).
In contrast to the previous studies that have used a key press response, Burton, Jenkins, and McNeill (2002) compared response latencies using a voice response for both the name and the semantic task. They used a set of just four famous faces, comprising two politicians and two pop stars where one person from each category was called Peter and the other one was called Paul. Despite practise at saying the names and occupations and repeated experimental trials, participants still took significantly longer to name these faces than to say the person's occupation. A different group of participants were asked to read aloud the printed words Peter, Paul, politician and pop star and showed no difference in articulation latencies when reading these words.
Difficulties in name recall have also been observed in neuropsychological studies where patients are able to recall the occupations but not the names of famous faces (e.g. Flude, Ellis, & Kay, 1989) and in learning studies where it is harder to learn someone's name than their occupation (e.g. Carson et al., 2000; Cohen & Faulkner, 1986; Stanhope & Cohen, 1993). This is the case even when the same word is used as an occupation and as a name (McWeeny, Young, Hay, & Ellis, 1987). Thus, it is harder to learn that someone is called Mr Baker than it is to learn that he is a baker.
Name retrieval difficulties can be explained by both serial and parallel models of face and person recognition. Bruce and Young's (1986) serial model of face recognition accounts for difficulties in name retrieval by suggesting that names are stored separately from other semantic information and that names can only be retrieved after some semantic information has been accessed. In contrast, interactive activation and competition (IAC) models of person recognition (e.g. Bredart, Valentine, Calder, & Gassi, 1995; Burton & Bruce, 1992) allow name and semantic information to be retrieved in parallel. Such models comprise pools of processing units, each corresponding to a proposition about a person (e.g. their name or nationality). Activation flows through these units in a continuous (rather than a staged) manner, and according to patterns of connectivity, different units can accrue differing amounts of activation over time. The details of how name disadvantages emerge are different between models, but the important thing to note here is that such accounts do not rely on strictly sequential access to information. Serial accounts such as Bruce and Young's propose that access to names can only occur following access to other information, and that the reverse pattern should never be observed. In contrast, faster name retrieval can be accounted for by parallel accounts of person recognition. In this regard, the evidence from studies of children is interesting because, to date, the only report of an advantage for name retrieval over retrieval of other personal information has been with children.
Children's face naming
Very little research has been carried out investigating children's naming of familiar faces. The primary reason for this is that it is very difficult to find a set of faces that a wide age range of children will know well. However, the studies that have been carried out suggest that there may be a difference between children and adults when it comes to name retrieval. Scanlan and Johnston (1997) carried out a matching task with 7-, 9-, 12- and 15-year-olds. A famous face was presented on the computer screen, followed by a name, an occupation or a nationality. Adults were faster to verify someone's occupation or nationality than their name, but 7- and 9-year-olds responded to names faster than occupations and nationality decisions. In addition, 12-year-olds showed no differences in RTs between name, nationality and occupation decisions, and only the 15-year-olds showed the same pattern as adults, whereby names were matched slower than occupations or nationalities.
In a second experiment, participants were required to name or give someone's occupation vocally on presentation of a face. Here, 7- and 9-year-olds named the faces faster than they were able to provide the person's occupation, whereas adults were slower at naming the faces than saying the person's occupation. There was no difference in response latencies between the three age groups when naming famous faces, but adults were much faster than both child groups when producing occupations. This result suggests that the naming advantage found in children in these two experiments may reflect children's difficulties with the semantic category relative to adults, rather than a genuine difference in the ways that children and adults process information about familiar faces.
Such a proposal is supported by recent work by Abdel Rahman, Sommer, and Olada (2004). They used a set of cartoon characters as the familiar stimuli, and the semantic knowledge associated with these characters was highly familiar to children. They found that both children and adults were faster at classifying cartoon characters on semantic decisions than when deciding if the character's name was made up of one or two words. They argue that when children are highly familiar with the semantic information associated with the familiar faces, then they are slower to respond to names like adults. This argument was further supported when children were found to be slower at classifying a set of learned faces on a semantic...
|
