Distinct roles for dorsal CA3 and CA1 in memory for sequential nonspatial events

Previous studies have suggested that dorsal hippocampal areas CA3 and CA1 are both involved in representing sequences of events that compose unique episodes. However, it is uncertain whether the contribution of CA3 is restricted to spatial information, and it is unclear whether CA1 encodes order per se or contributes by an active maintenance of memories of sequential events. Here, we developed a new behavioral task that examines memory for the order of sequential nonspatial events presented as trial-unique odor pairings. When the interval between odors within a studied pair was brief (3 sec), bilateral dorsal CA3 lesions severely disrupted memory for their order, whereas dorsal CA1 lesions did not affect performance. However, when the inter-item interval was extended to 10 sec, CA1 lesions, as well as CA3 lesions, severely disrupted performance. These findings suggest that the role of CA3 in sequence memory is not limited to spatial information, but rather appears to be a fundamental property of CA3 function. In contrast, CA1 becomes involved when memories for events must be held or sequenced over long intervals. Thus, CA3 and CA1 are both involved in memory for sequential nonspatial events that compose unique experiences, and these areas play different roles that are distinguished by the duration of time that must be bridged between key events.Episodic memory involves the ability to encode and retrieve the order of events in individual experiences (Tulving 1983). Recent evidence in both animals and humans indicates that the hippocampus plays a critical role in this capacity. In animals, damage to the hippocampus impairs memory for the order of associated elements that compose an episode (Fortin et al. 2002; Kesner et al. 2002), and hippocampal neuronal activity reflects processing of the order of events in both spatial (Dragoi and Buzsáki 2006; Foster and Wilson 2007) and nonspatial episodes (Manns et al. 2007). In humans, hippocampal activation has also been related to memory for the order of elements (Kumaran and Maguire 2006; Lehn et al. 2009; Ross et al. 2009).Within the hippocampal circuitry, contributions of the CA3 and CA1 fields are probably most extensively studied, but this work has not yet clarified the distinct roles of these areas in sequence memory. Computational models suggest that the recurrent connections of CA3 cells operate as an attractor network that computes associations between elements (Norman and O''Reilly 2003; Rolls 2007) and is suitable for representing sequences of events in episodic memories (Jensen and Lisman 1996; Levy 1996; Lisman 1999). Studies on the effects of selective damage within the hippocampus have shown that CA3 is critical for remembering sequences of spatial locations (Hunsaker et al. 2008a), but not sequences of nonspatial events (Hoge and Kesner 2007). It is, therefore, uncertain whether CA3 is critical for sequence memory per se, rather than other aspects of spatial processing. Other observations suggest that CA1 may be involved in memory for the order of both spatial (Hunsaker et al. 2008a) and nonspatial stimuli (Hoge and Kesner 2007; Manns et al. 2007). However, it is not clear whether the contribution of CA1 involves integrating sequential elements of a memory or instead participates by active maintenance of event memories that underlies bridging sequential events in an episode (Kesner et al. 2005).To shed light on these issues, we compared the effects of selective damage to CA3 and CA1 on memory for the order of nonspatial events that occurred in unique episodes. We designed a task, based on the delayed-nonmatching-to-sample test, wherein subjects were required to remember the order of two sequentially presented stimuli in trial–unique-paired associations (Fig. 1).Open in a separate windowFigure 1.Test of memory for the order of stimuli in trial-unique odor pairs. At study, animals were presented with 10 odor-paired associates and odors in a pair were presented one at a time. At test, animals were presented with the same 10 odor pairs and were required to distinguish pairs where the odors within a pair were presented in the same order as during study (“old”) from pairs where the odors were presented in the reverse order (“new”). Old and new order test pairs were presented in a pseudorandom order. The first odor in each test pair acted as a cue to the ordering of the odors within a test pair; the animal was required to place its nose over the cup, but no digging response was required or rewarded. When the second cup was presented, the animal could dig to retrieve a reward if the order was new. If the order was old, the animal was required to approach an empty cup in the back of the home cage to obtain reward.