Limits...
Impaired spatial and non-spatial configural learning in patients with hippocampal pathology.

Kumaran D, Hassabis D, Spiers HJ, Vann SD, Vargha-Khadem F, Maguire EA - Neuropsychologia (2007)

Bottom Line: Our data also provide evidence that residual configural learning can occur in the presence of significant hippocampal dysfunction.Moreover, evidence obtained from a post-experimental debriefing session suggested that patients acquired declarative knowledge of the underlying task contingencies that corresponded to the best-fit strategy identified by our strategy analysis.In summary, our findings support the notion that the hippocampus plays an important role in both spatial and non-spatial configural learning, and provide insights into the role of the medial temporal lobe (MTL) more generally in incremental reinforcement-driven learning.

View Article: PubMed Central - PubMed

Affiliation: Wellcome Trust Centre for Neuroimaging, Institute of Neurology, University College London, 12 Queen Square, London WC1N 3BG, UK. d.kumaran@fil.ion.ucl.ac.uk

ABSTRACT
The hippocampus has been proposed to play a critical role in memory through its unique ability to bind together the disparate elements of an experience. This hypothesis has been widely examined in rodents using a class of tasks known as "configural" or "non-linear", where outcomes are determined by specific combinations of elements, rather than any single element alone. On the basis of equivocal evidence that hippocampal lesions impair performance on non-spatial configural tasks, it has been proposed that the hippocampus may only be critical for spatial configural learning. Surprisingly few studies in humans have examined the role of the hippocampus in solving configural problems. In particular, no previous study has directly assessed the human hippocampal contribution to non-spatial and spatial configural learning, the focus of the current study. Our results show that patients with primary damage to the hippocampus bilaterally were similarly impaired at configural learning within both spatial and non-spatial domains. Our data also provide evidence that residual configural learning can occur in the presence of significant hippocampal dysfunction. Moreover, evidence obtained from a post-experimental debriefing session suggested that patients acquired declarative knowledge of the underlying task contingencies that corresponded to the best-fit strategy identified by our strategy analysis. In summary, our findings support the notion that the hippocampus plays an important role in both spatial and non-spatial configural learning, and provide insights into the role of the medial temporal lobe (MTL) more generally in incremental reinforcement-driven learning.

Show MeSH

Related in: MedlinePlus

Experimental design: subjects were instructed to play the role of a weather forecaster, and try to learn over the course of the experiment how different “patterns” of shapes on the screen were associated with one of two outcomes, sun or rain (see Section 2). Each one of the eight patterns was associated with an outcome in a deterministic fashion (i.e. with 100% probability). In patterns 1–4, the position of the triangle determines the outcome (in this example, although the allocation of shapes to outcomes was changed between subjects). Hence when the triangle appears on the left, the outcome is sun regardless of the shape present in the centre. When the triangle appears on the right, the outcome is always rain. In patterns 5–8, specific shape–shape pairings determine the outcome, with the position of the square being irrelevant. Hence, square together with star is associated with sun, regardless of the position of the square. Conversely, square together with ellipse is always associated with rain. Trials could therefore be divided conceptually into those involving learning of spatial (patterns 1–4), as opposed to non-spatial (patterns 5–8), configural associative information.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2572909&req=5

fig1: Experimental design: subjects were instructed to play the role of a weather forecaster, and try to learn over the course of the experiment how different “patterns” of shapes on the screen were associated with one of two outcomes, sun or rain (see Section 2). Each one of the eight patterns was associated with an outcome in a deterministic fashion (i.e. with 100% probability). In patterns 1–4, the position of the triangle determines the outcome (in this example, although the allocation of shapes to outcomes was changed between subjects). Hence when the triangle appears on the left, the outcome is sun regardless of the shape present in the centre. When the triangle appears on the right, the outcome is always rain. In patterns 5–8, specific shape–shape pairings determine the outcome, with the position of the square being irrelevant. Hence, square together with star is associated with sun, regardless of the position of the square. Conversely, square together with ellipse is always associated with rain. Trials could therefore be divided conceptually into those involving learning of spatial (patterns 1–4), as opposed to non-spatial (patterns 5–8), configural associative information.

Mentions: In this study, we explored the role of the human hippocampus in memory for spatial and non-spatial configural associative information in the setting of a novel associative learning task consisting of multiple trials with feedback. Subjects were instructed to play the role of a weather forecaster, and try to learn over the course of the experiment how different patterns of shapes on the screen were associated with one of two outcomes i.e. sun or rain (see Section 2). Each one of eight patterns (Fig. 1) was associated with an outcome in a deterministic fashion (i.e. with 100% probability). As is evident from the upper four patterns (1–4) in Fig. 1 the position of the triangle determines the outcome (in this example, although the allocation of shapes to outcomes was changed between subjects). Therefore, when the triangle appears on the left, the outcome is sun regardless of the shape present in the centre. Conversely, when the triangle appears on the right, the outcome is always rain. From the bottom four patterns (Fig. 1: patterns 5–8), it is evident that it is the specific shape–shape pairings that determine the outcome, with the position of the square being irrelevant. Square together with star is associated with sun, regardless of the position of the square. Conversely, square together with ellipse is always associated with rain. Hence, although all eight patterns were intermixed pseudorandomly throughout the experiment (see Section 2), trials could be divided conceptually into those involving learning of spatial (Fig. 1: patterns 1–4), as opposed to non-spatial (Fig. 1: patterns 5–8), configural associative information. Critically, subjects could only solve the task by learning the outcomes associated with specific shape–location (i.e. spatial configural) and shape–shape (i.e. non-spatial configural) pairings. As such, subjects could not solve the task by using only elemental information (i.e. a single shape or position), and only achieve a maximum of 75% correct responses using this strategy. We also employed a strategy analysis that permitted us to characterize the nature of the information (i.e. elemental versus configural associative) acquired by subjects during learning (see Section 2). Of note, our task, although superficially similar to the “standard” weather prediction task (Poldrack et al., 2001; Hopkins, Myers, Shohamy, Grossman, & Gluck, 2004; Knowlton, Mangels, & Squire, 1996), differs considerably since it involves the learning of configural (as opposed to elemental) information and is deterministic (rather than probabilistic) in nature.


Impaired spatial and non-spatial configural learning in patients with hippocampal pathology.

Kumaran D, Hassabis D, Spiers HJ, Vann SD, Vargha-Khadem F, Maguire EA - Neuropsychologia (2007)

Experimental design: subjects were instructed to play the role of a weather forecaster, and try to learn over the course of the experiment how different “patterns” of shapes on the screen were associated with one of two outcomes, sun or rain (see Section 2). Each one of the eight patterns was associated with an outcome in a deterministic fashion (i.e. with 100% probability). In patterns 1–4, the position of the triangle determines the outcome (in this example, although the allocation of shapes to outcomes was changed between subjects). Hence when the triangle appears on the left, the outcome is sun regardless of the shape present in the centre. When the triangle appears on the right, the outcome is always rain. In patterns 5–8, specific shape–shape pairings determine the outcome, with the position of the square being irrelevant. Hence, square together with star is associated with sun, regardless of the position of the square. Conversely, square together with ellipse is always associated with rain. Trials could therefore be divided conceptually into those involving learning of spatial (patterns 1–4), as opposed to non-spatial (patterns 5–8), configural associative information.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2572909&req=5

fig1: Experimental design: subjects were instructed to play the role of a weather forecaster, and try to learn over the course of the experiment how different “patterns” of shapes on the screen were associated with one of two outcomes, sun or rain (see Section 2). Each one of the eight patterns was associated with an outcome in a deterministic fashion (i.e. with 100% probability). In patterns 1–4, the position of the triangle determines the outcome (in this example, although the allocation of shapes to outcomes was changed between subjects). Hence when the triangle appears on the left, the outcome is sun regardless of the shape present in the centre. When the triangle appears on the right, the outcome is always rain. In patterns 5–8, specific shape–shape pairings determine the outcome, with the position of the square being irrelevant. Hence, square together with star is associated with sun, regardless of the position of the square. Conversely, square together with ellipse is always associated with rain. Trials could therefore be divided conceptually into those involving learning of spatial (patterns 1–4), as opposed to non-spatial (patterns 5–8), configural associative information.
Mentions: In this study, we explored the role of the human hippocampus in memory for spatial and non-spatial configural associative information in the setting of a novel associative learning task consisting of multiple trials with feedback. Subjects were instructed to play the role of a weather forecaster, and try to learn over the course of the experiment how different patterns of shapes on the screen were associated with one of two outcomes i.e. sun or rain (see Section 2). Each one of eight patterns (Fig. 1) was associated with an outcome in a deterministic fashion (i.e. with 100% probability). As is evident from the upper four patterns (1–4) in Fig. 1 the position of the triangle determines the outcome (in this example, although the allocation of shapes to outcomes was changed between subjects). Therefore, when the triangle appears on the left, the outcome is sun regardless of the shape present in the centre. Conversely, when the triangle appears on the right, the outcome is always rain. From the bottom four patterns (Fig. 1: patterns 5–8), it is evident that it is the specific shape–shape pairings that determine the outcome, with the position of the square being irrelevant. Square together with star is associated with sun, regardless of the position of the square. Conversely, square together with ellipse is always associated with rain. Hence, although all eight patterns were intermixed pseudorandomly throughout the experiment (see Section 2), trials could be divided conceptually into those involving learning of spatial (Fig. 1: patterns 1–4), as opposed to non-spatial (Fig. 1: patterns 5–8), configural associative information. Critically, subjects could only solve the task by learning the outcomes associated with specific shape–location (i.e. spatial configural) and shape–shape (i.e. non-spatial configural) pairings. As such, subjects could not solve the task by using only elemental information (i.e. a single shape or position), and only achieve a maximum of 75% correct responses using this strategy. We also employed a strategy analysis that permitted us to characterize the nature of the information (i.e. elemental versus configural associative) acquired by subjects during learning (see Section 2). Of note, our task, although superficially similar to the “standard” weather prediction task (Poldrack et al., 2001; Hopkins, Myers, Shohamy, Grossman, & Gluck, 2004; Knowlton, Mangels, & Squire, 1996), differs considerably since it involves the learning of configural (as opposed to elemental) information and is deterministic (rather than probabilistic) in nature.

Bottom Line: Our data also provide evidence that residual configural learning can occur in the presence of significant hippocampal dysfunction.Moreover, evidence obtained from a post-experimental debriefing session suggested that patients acquired declarative knowledge of the underlying task contingencies that corresponded to the best-fit strategy identified by our strategy analysis.In summary, our findings support the notion that the hippocampus plays an important role in both spatial and non-spatial configural learning, and provide insights into the role of the medial temporal lobe (MTL) more generally in incremental reinforcement-driven learning.

View Article: PubMed Central - PubMed

Affiliation: Wellcome Trust Centre for Neuroimaging, Institute of Neurology, University College London, 12 Queen Square, London WC1N 3BG, UK. d.kumaran@fil.ion.ucl.ac.uk

ABSTRACT
The hippocampus has been proposed to play a critical role in memory through its unique ability to bind together the disparate elements of an experience. This hypothesis has been widely examined in rodents using a class of tasks known as "configural" or "non-linear", where outcomes are determined by specific combinations of elements, rather than any single element alone. On the basis of equivocal evidence that hippocampal lesions impair performance on non-spatial configural tasks, it has been proposed that the hippocampus may only be critical for spatial configural learning. Surprisingly few studies in humans have examined the role of the hippocampus in solving configural problems. In particular, no previous study has directly assessed the human hippocampal contribution to non-spatial and spatial configural learning, the focus of the current study. Our results show that patients with primary damage to the hippocampus bilaterally were similarly impaired at configural learning within both spatial and non-spatial domains. Our data also provide evidence that residual configural learning can occur in the presence of significant hippocampal dysfunction. Moreover, evidence obtained from a post-experimental debriefing session suggested that patients acquired declarative knowledge of the underlying task contingencies that corresponded to the best-fit strategy identified by our strategy analysis. In summary, our findings support the notion that the hippocampus plays an important role in both spatial and non-spatial configural learning, and provide insights into the role of the medial temporal lobe (MTL) more generally in incremental reinforcement-driven learning.

Show MeSH
Related in: MedlinePlus