Ordering and serial learning

The ability to put objects in order and to memorise them as a whole sequence is crucial for an intelligent action. Serial memory is the ability to encode and retrieve a list of items in their correct temporal order. The nature of the mental representation that allows to retrieve a list of items is still unclear. As many other approaches for studying animal intelligence, the experimental paradigm of what is also called “ list learning ” is derived from psychological studies, namely, from the classic experiments of Ebbinghaus (1885; see Chapter 3) and Ebenholtz (1963), who first investigated the organisation of such sequences in experiments on the memorisation of nonsense syllables. Terrace (2001) suggests considering this classic paradigm together with the mastery of various types of mazes (Small, 1900; see Chapter 1) because the results of both types of experiments gave rise to the classic theory that states that serially organised behaviour can be represented as a linear sequence of associations. Ebbinghaus explained list learning by reference to associations between successive items and between a particular item and its list position. Hull offered a similar explanation of maze learning by rats (Hull, 1943). Thus, associative principles that were used to explain how a human adult memorises a list of arbitrary items were used to explain how an experimentally naive rat learns a sequence of arbitrary responses, and vice versa (Osgood, 1953; Terrace, 2001).

Lashley (1951) rejected linear models of serially organised behaviour because they could not explain the knowledge of relationships between non-adjacent items. Recently cognitive ethologists and neurophysiologists have raised a question of serially organised association by augmenting linear structures of stimuli with hierarchical structures (Orlov et al., 2002). This problem is based on the concept of chunking that was first introduced by Miller (1956) in his paper " On the magical number 7". Miller argued that a chunk was the basic unit for measuring the capacity of immediate memory (in current terminology, short-term or working memory; see Part III). The idea was that subjects could retain a large number of discrete items of information if they were encoded as chunks before they were transferred to long-term memory. For example, the 12 digits 1-4-9-2-1-7-7-6-1-8-1-2 could be encoded as 3 historical dates. In contrast to the enormous capacity of long-term memory (LTM), Miller estimated the capacity of STM to be 7 ± 2 chunks and argued that the amount of information that is retained in STM is independent of the amount of information contained by each chunk.

One of the most relevant methods that allow studying serial learning in non-verbal subjects is the so-called simultaneous chaining procedure that differs from those used in previous studies of serial learning in animals (Straub, Terrace, 1981; Terrace, 1987; Chen et al., 1997; Terrace et al., 2003). The main idea of this method is that a subject has not to simply recognise learned stimuli (multiple matching-to samples), but to choose the stimuli in a definite sequential order. Unlike the successive chaining paradigm, a simultaneous chaining paradigm presents all list items throughout each trial (e.g., the numbers on the face of a telephone). In a successive chain, the subject encounters each cue individually (e.g., the choice points in a maze). A second difference is the variation of the physical configuration of list items from trial to trial. This prevents subjects from using a particular physical sequence of responses to produce the required list (for example, when making a telephone call with a sequence of learned movements on a number pad). To execute a simultaneous chain correctly, the subject has to respond to each item in a particular order, regardless of its spatial position.

Experiments on monkeys and pigeons showed that they learn sequences of stimuli much more readily in a case the stimuli could be clustered into groups. Straub et al. (1979) trained pigeons to respond to randomly configured arrays of four colours in a particular sequence: red –green-yellow-blue. Pigeons learned the 4-item list of colours. The extensive training time that pigeons need to master a 4-item list (more than 3 months) suggests that 4 items may approach the limit of their memory span. For human subjects the classic remedy for overcoming limitations of memory span is to reorganise unrelated list items into chunks (Miller, 1956). The efficacy of that approach was evaluated with pigeons that were trained to learn 5-item lists composed of colours and achromatic geometric forms. When a number of items increased, pigeons responded at chance levels of accuracy to subsets drawn from lists on which items were not clustered. Pigeons showed no signs of improvement on successive 3- or 4-item lists, each composed of novel items (colour photographs of natural scenes).

D'Amato and Colombo (1988) used the simultaneous chaining procedure to train capuchin monkeys to produce arbitrary 4-item lists. Monkeys were trained on four lists, each containing four novel photographs of natural objects (flowers, fruits, animals). The task was to touch the simultaneously presented images in the correct order (A1– A2–A3–A4, B1–B2– B3–B4, C1–C2–C3–C4, D1–D2–D3–D4). When the monkeys had mastered this task, the items were shuffled, taking one item from each list, so that in two derived lists the ordinal numbers of the items were maintained (e.g. A1–D2–C3–B4) while in two others they were not (e.g. B3– A1–D4–C2). Lists with maintained ordinal position were acquired rapidly and virtually without error, while derived lists in which the ordinal position was changed were as difficult to learn as novel lists. This pattern of transfer to derived lists implies that the monkeys originally acquired some knowledge about each item's ordinal position, rather than only generating a chain of serial pair associations for each list of items. Monkeys acquired 5-item lists more rapidly than pigeons. Pigeons showed no signs of improvement on successive 3- or 4-item lists, each composed of novel items (colour photographs of natural scenes). Another important difference between the serial skills of monkeys and

pigeons was the ease of acquiring new lists. Monkeys trained to learn successive 4- and 6-item lists of different photographs became progressively more efficient at mastering each list (Swartz et al., 1991; Chen et al., 1997). More recently Terrace (2001) showed that rhesus monkeys can learn 7-item lists . Monkeys were first trained on 3- and 4-item lists. The subjects were then trained in the same manner on four 7-item lists. The monkeys not only mastered each list but they did so with progressively fewer trials on each new list. To place this achievement in a perspective, the probability of guessing correctly the ordinal position of each item at the start of training on a 7-item list is 1/7! <.0002.

Another line of research addresses what is called serial recognition, that is, the ability of monkeys to recognise stimuli that are presented sequentially. In the first experiment, subjects were shown a sequence of novel photographs at the beginning of each trial. They were then required to select these photographs from a simultaneous display that included “distractors”. Subjects were allowed to respond to the items in the display in any order they chose. To obtain a reward they had to select all of the items shown at the beginning of the trial and avoid responding to any of the distractors. Subjects were able to recognise all items of 4-item sequences at high levels of accuracy even when those items were embedded in displays containing 5 distractors. Because there are no constraints on the order in which items could be selected, this paradigm is methodologically similar to free-recall studies with human subjects. One empirical similarity between the recall of sequentially presented items by monkeys and humans is the lengthening of the response time for each successive item that is reported. The monotonically increasing function relating response time to the order of reporting appears to reflect the difficulty of searching the contents of short-term memory as more items are recalled rather than the time needed for a visual search of the actual display (Terrace, 2001).