Elsevier

Behavioural Brain Research

Volume 201, Issue 2, 12 August 2009, Pages 285-291
Behavioural Brain Research

Research report
Spatial reorientation in rats (Rattus norvegicus): Use of geometric and featural information as a function of arena size and feature location

https://doi.org/10.1016/j.bbr.2009.02.026Get rights and content

Abstract

Rats were used in a spatial reorientation task to assess their ability to use geometric and non-geometric, featural, information. Experimental conditions differed in the size of the arena (small, medium, or large) and whether the food-baited corner was near or far from a visual feature. The main measure was the percentage of trials with first-choice visits to the baited corner (Corner C) and the geometric equivalent corner (Corner R). Regardless of arena size, only the rats in the near-feature condition learned to make more first-choice visits to Corner C than Corner R. However, in this condition, there was a significant trend suggesting an increasing and decreasing use of, respectively, featural and geometric information with increasing arena size. Continued training with the feature removed caused all rats to primarily visit Corners C and R, reflecting the exclusive use of geometric information. However, again, there was a significant trend suggesting a decreasing use of this type of information with increasing arena size. These results were discussed in the framework of data from previous studies assessing reorientation in other species and of an associative-learning model.

Introduction

Spatial representations enable animals to orient in their own environment. A spatial reorientation paradigm helps us to understand the nature of the environmental features serving as the constitutive elements of spatial representations, allowing spatial mapping processes. The ability to reorient is the result of one's capacity of identifying a location and of establishing a heading by reference to the characteristics of the external environment [6]. Thus, animals in the absence of (very) distant external reference cues, such as the sun, must resort to more local spatial information. Experiments with several animal species, such as fishes (Xenotoca eiseni, Carassius auratus), chicks (Gallus gallus), pigeons (Columba livia), rats (Rattus norvegicus), rhesus monkeys (Macaca mulatta), and humans have been performed in order to assess which kind of spatial information disoriented animals use for reorienting, both from a comparative and a developmental perspective (see for a review [3]). All these studies focused on the encoding of geometric and non-geometric environmental information.

A typical example of an experimental set-up used in the corresponding research is shown in Fig. 1a. Specifically, one corner of a rectangular enclosure is defined as the ‘correct’ Corner (C), either by instruction or by providing a reward for visits to it. In one experimental version, in a so-called reference-memory paradigm, the location of the correct corner remains the same from trial to trial and the subject is required to remember and relocate this corner after disorientation. Disorientation may involve rotation of the experimental subject or removal from the enclosure, followed by replacement in a randomly chosen location. In case the arena does not contain any specific ‘featural’ information, that is, when in Fig. 1a all walls have the same colour, a reliable finding is that, upon replacement into the enclosure, experimental subjects visit Corner C, and the corner that is geometrically identical to this corner, namely Corner R (at 180° rotation), an equal number of times. Moreover, the subjects visit each of these corners significantly more than either the corner nearest to Corner C (Corner NC), or the corner that is nearest to Corner R (Corner NR). This pattern of results is suggestive of the use of geometrical information to reorient. The ability to use this type of information seems to be widespread among vertebrates.

Correct relocation of Corner C may be possible on the basis of additional non-geometric or ‘featural’ information. This type of information may be provided by a specific landmark located near Corner C, such as when the short wall adjacent to such corner is of a different colour compared to the other walls (see Fig. 1a). Corner C can now be unequivocally distinguished from Corner R.

Results of studies examining the ability of various species to successfully locate Corner C in an experimental set-up as described above, implying the capacity to conjoin geometric and non-geometric (featural) information, are mixed. Three fish species have been shown to be able to combine geometric information and featural information provided by a coloured wall (referring to Fig. 1a: either the top long wall or the right short wall; redtail splitfins: [16], [18]; goldfish: [23]; convict fish: [1]). The fish is also able to use featural information consisting of panels located in the corners. However, when these panels are only present far from the target, the animals fail to combine the information provided by the remaining features and the geometry [17].

Chicks have also been found able to conjoin geometric and featural information provided by visually distinct panels located in the corners. However, as in fish, this capacity seems to be lost when the featural information solely consists of panels in Corners NC and NR [20], [21].

Pigeons were shown able to conjointly use geometric and non-geometric information even when the latter (i.e., distinctive panels located in the corners) is presented far, rather than near, the target corner [10].

Rhesus monkeys have also been found able to combine the two types of information under certain conditions, that is, both when the featural information consisted of a coloured short wall near the target Corner C (see Fig. 1a) and when the short coloured wall was on the opposite side of the arena (i.e., near Corners R and NR in Fig. 1b). However, monkeys systematically make rotational errors (i.e., visits to Corner R) in both such conditions, and were not able to combine the two types of information when the distant feature consisted of distinctive panels [8].

Human children less than 5 years of age have been reported to be both able [11], [12] and unable (i.e., visiting Corners C and R equally often; [9], [11]) to use featural information provided by a coloured short wall (right wall as in Fig. 1a). This discrepancy may be related to differences in size of the environment. Specifically, use of non-geometric cues may be restricted to relatively large environments [11]. However, as in the monkey studies [8], even in large environments the children still made systematic rotational errors, visiting Corner R more often than each of the other incorrect corners.

Data for human adults largely correspond with those of older children: they are able to combine the two types of information regardless of the size of the environment and they make relatively few rotational errors (see Ref. [3] for reviews).

Overall, it seems clear that the spatial scale of the environment plays a crucial role in animals’ ability to conjoin geometric and non-geometric information (see for a review [5]). Fishes reoriented by conjoining geometric and non-geometric information independently of the size of the enclosure [18]. Moreover, they were able to reorient when relocated from a large to a small experimental space and vice versa. But an effect of the size was observed with respect to the type of errors fishes made [18], [20]: when a generalization occurred from the small to the large tank, they committed relatively more geometric errors, whereas when a generalization occurred from the large to the small tank, fishes committed relatively more non-geometric errors.

Also domestic chicks (Gallus gallus) conjoin geometric and non-geometric information in small and large enclosures [22]. Moreover, they reorient immediately when displaced from a large to a small space. Unlike fishes and toddlers, they did not show any differences in the amount of errors. With chicks, the effect of size emerged only when geometric and non-geometric cues provided contradictory information. This was experimentally obtained with transformations altering the geometric relation between the target and the shape of the environment. In such condition, when tested in a small enclosure, chicks relied on geometry, whereas, in the large enclosure, they relied on local features [22]. The claim that chicks encode both types of information in large as well as in small enclosures is further supported by more recent experiments [4]. In these experiments, differences were observed as a function of the size of the experimental space. After removal of the featural cue, chicks searched at the two geometrically equivalent corners, after removal of the geometrical cue they searched at the corner identified by the correct feature; use of residual geometrical information was stronger in small than large arenas whereas use of residual non-geometric information was stronger in large than in small arenas.

Finally, another animal species that has been examined, which is also the focus of attention in the present experiment, is the rat. Rats are able to combine geometric and non-geometric information (provided by landmarks) in a reference-memory task but not in a working-memory task [2], in an escape task but not in search task [7], although rotational errors are made relatively frequently. Moreover, when only ‘distant’ featural information is present in the form of landmarks in Corners NC and NR, rats make as many visits to Corner C as to Corner R [2].

These comparative data suggest two factors that may play a role in the ability to conjoin geometric and non-geometric information. The first factor is the location of the featural cue. Using only landmarks that were relatively remote from the target Corner C (namely in Corners NC and NR), fish, chicks, and rats appeared to be unable to use the featural information provided by these landmarks, whereas pigeons could use non-geometric information both when this was placed near or far from the target corner.

However, rats were able to do so when a feature was present near the target corner using as landmarks both distinctive panels located in the corners and a continuous surface (such as a coloured wall; see Fig. 1b). The second relevant factor, as mentioned above, is the size of the enclosure. At least in young human children (but not older children and adults), the featural information appears to be only used when the experimental room is relatively large [11]. In fishes and chicks, arena-size manipulations did not affect the ability to use featural information provided by a coloured wall even though subjects made different types of errors as a function of arena size, showing a preferential use of geometric information in small enclosures and non-geometrical information in large enclosures. It remains to be seen whether this is also true for rats.

The purpose of the present study was to further assess the conditions under which rats are able to use the information provided by geometric and featural cues. Specifically, rats were trained in experimental conditions that differed in arena size and/or in the location of featural information that was provided by the colour of a short wall (either the short wall near Corners C and NC or Corners NR and R). If size of the enclosure is an important determinant of the ability to use the featural information, as it appears to be the case in human children, rats should only be able to distinguish between Corners C and R in relatively large enclosures. However, even in such a large arena, this ability may be restricted to the condition in which the feature is located near the target corner.

Section snippets

Subjects

The subjects were 12 male Wistar rats, bred and reared at the Radboud University Nijmegen, The Netherlands. The animals were experimentally naïve and had a mean body weight of 396 g prior to the experiment. The rats were housed in Plexiglas cages with a bedding of wood chips. They had free access to water, whereas availability of food was restricted, to maintain the animals at 85% of their free-feeding weight. The rats were maintained on a 12/12-h dark/light cycle, with all experimental sessions

Training with feature

The rats in each arena-size and feature-location condition showed a gradual decrease across sessions in the mean number of corners visited before eating two pellets. The overall mean number of visits on each trial, pooled across arena size and feature location, decreased from 4.38 to 2.25 from Session 1 to Session 16. Arena (arena size: 3 levels) × Feature Location (2 levels: near or far from target corner) × Session (16 levels) ANOVA using the mean number of visits only revealed a significant main

Discussion

The data of the first training phase suggest that the rats were able to use geometric information and featural information given by a coloured wall, provided the latter is presented near the target location. Moreover, the larger the arena, the more the rats were attracted by the feature cue, and the less they used geometrical information. After removal of the white wall, the rats primarily used geometrical information, supporting the claim that during the first training phase with the feature

References (23)

  • C.R. Gallistel

    The organization of learning

    (1990)
  • Cited by (16)

    • Size does not matter, but features do: Clark's nutcrackers (Nucifraga columbiana) weigh features more heavily than geometry in large and small enclosures

      2014, Behavioural Processes
      Citation Excerpt :

      Over the years, many species have been shown to incidentally encode surface geometry when reorienting, and that featural and geometric cues are integrated when forming a spatial representation of the environment (see Cheng, 2008; Cheng et al., 2013; Cheng and Newcombe, 2005). Recently, researchers have also shown that the size of the search space may be important for the relative weighing of featural and geometric information (e.g., chicks: Chiandetti et al., 2007; Chiandetti and Vallortigara, 2008; Vallortigara et al., 2005; fish: Sovrano et al., 2005, 2007; humans: Learmonth et al., 2002, 2008; Ratliff and Newcombe, 2008; Sturz and Kelly, 2013; rats: Maes et al., 2009). Hermer and Spelke (1994) reported that human toddlers did not encode featural cues when examined using a reference memory task similar to Cheng (1986).

    • Hierarchical use of cues in the missing object recognition task by rats (Rattus norvegicus)

      2013, Behavioural Processes
      Citation Excerpt :

      Field experiments on the non-caching rufous humming bird, Selasphorus rufus (Healy and Hurly, 1998; Hurly and Healy, 1996) demonstrate that these birds switch from relying more on the local to the global positions of previously sampled ‘flowers’ within their arrays when they become separated from each other by more than 40 cm. Rats also reduce their use of the geometry of a rectangular enclosure and increase their use of their corner positions of cues to find hidden food as a rectangular enclosure is enlarged (Maes et al., 2009). Perhaps rats in our study would have also preferred to use or retrieve the correct feeder's global position earlier during their choice sequence in arrays of more widely separated feeders.

    • Environment, cognition, and culture: Reconsidering the cognitive map

      2013, Journal of Environmental Psychology
      Citation Excerpt :

      The size of the testing enclosure was also found to affect whether adults utilized geometric or feature information (Ratliff & Newcombe, 2008). Likewise, room size has been shown to increase the use of feature information as compared to the geometry of the enclosure in studies with chicks (Chiandetti et al., 2007) and with rats (Maes, Fontanari, & Regolin, 2009). Findings such as these are consistent with Cheng's (2008) later reassessment, claiming that use of geometric information was not “obligatory” across species.

    • From natural geometry to spatial cognition

      2012, Neuroscience and Biobehavioral Reviews
      Citation Excerpt :

      Basically, it seems that for redtail splitfins geometric information is relatively more important than featural information than it is for chicks. Perhaps such a difference could be expected considering that birds are highly visual animals with considerable spatial resolution capabilities; fish, in contrast, because of adaptation to an aquatic environment, show comparatively more reduced spatial resolution because of the spatial filtering produced by water [but see recent evidences collected with rats (R. norvegicus) which used more featural than geometric information with increasing arena size (Maes et al., 2009)]. Reorientation in an enclosed environment is a task that depends on the shape of the search space in order to obtain the directional information needed to localize a goal.

    View all citing articles on Scopus
    View full text