Cephalopods are sensitive to the linear polarization characteristics of light. To examine if this polarization sensitivity plays a role in the predatory behavior of cuttlefish, we examined the preference of Sepia officinalis when presented with fish whose polarization reflection was greatly reduced versus fish whose polarization reflection was not affected. Cuttlefish preyed preferably on fish with normal polarization reflection over fish that did not reflect linearly polarized light (n = 24, chi 2 = 17.3, P < 0.0001), implying that polarization sensitivity is used during predation. We suggest that polarization vision is used to break the countershading camouflage of light-reflecting silvery fish.

No abstract

Several studies have demonstrated that mammals, birds and fish use comparable spatial learning strategies. Unfortunately, except in insects, few studies have investigated spatial learning mechanisms in invertebrates. Our study aimed to identify the strategies used by cuttlefish (Sepia officinalis) to solve a spatial task commonly used with vertebrates. A new spatial learning procedure using a T-maze was designed. In this maze, the cuttlefish learned how to enter a dark and sandy compartment. A preliminary test confirmed that individual cuttlefish showed an untrained side-turning preference (preference for turning right or left) in the T-maze. This preference could be reliably detected in a single probe trial. In the following two experiments, each individual was trained to enter the compartment opposite to its side-turning preference. In Experiment 1, distal visual cues were provided around the maze. In Experiment 2, the T-maze was surrounded by curtains and two proximal visual cues were provided above the apparatus. In both experiments, after acquisition, strategies used by cuttlefish to orient in the T-maze were tested by creating a conflict between the formerly rewarded algorithmic behaviour (turn, response learning) and the visual cues identifying the goal (place learning). Most cuttlefish relied on response learning in Experiment 1; the two strategies were used equally often in Experiment 2. In these experiments, the salience of cues provided during the experiment determined whether cuttlefish used response or place learning to solve this spatial task. Our study demonstrates for the first time the presence of multiple spatial strategies in cuttlefish that appear to closely parallel those described in vertebrates.

Octopuses forage far from temporary home dens to which they return for shelter. Spatial tasks may assess learning. Octopuses (Octopus bimaculoides) were placed in a novel arena, and their movements were tracked for 72 hr. Movements around the arena decreased across time, consistent with exploratory learning. Next, octopuses were given 23 hr to move around an arena; after a 24-hr delay, their memory of a burrow location was tested. Most remembered the location of the open burrow, demonstrating learning in 1 day. Finally, octopuses were trained to locate a single open escape burrow among 6 possible locations. Retention was tested after a week and was immediately followed by reversal training (location rotated 180 degrees ). Octopuses learned the original location of the burrow, remembering it for a week. Path lengths increased significantly after reversal, gradually improving and showing relearning. Octopuses show exploratory behavior, learning, and retention of spatial information.

Laboratory mazes were used to study spatial-learning capabilities in cuttlefish (Sepia officinalis), using escape for reinforcement. In preliminary observations, cuttlefish in an artificial pond moved actively around the environment and appeared to learn about features of their environment. In laboratory experiments, cuttlefish exited a simple alley maze more quickly with experience and retained the learned information. Similar improvement was not found in open-field mazes or T mazes, perhaps because of motor problems. Cuttlefish learned to exit a maze that required them to find openings in a vertical wall. The wall maze was modified to an arena, and simultaneous discrimination learning and reversal learning were demonstrated. These experiments indicate that cuttlefish improve performance over serial reversals of a simultaneous, visual-spatial discrimination problem.

Sleep has been observed in several invertebrate species, but its presence in marine invertebrates is relatively unexplored. Rapid-eye-movement (REM) sleep has only been observed in vertebrates. We investigated whether the cuttlefish Sepia officinalis displays sleep-like states. We find that cuttlefish exhibit frequent quiescent periods that are homeostatically regulated, satisfying two criteria for sleep. In addition, cuttlefish transiently display a quiescent state with rapid eye movements, changes in body coloration and twitching of the arms, that is possibly analogous to REM sleep. Our findings thus suggest that at least two different sleep-like states may exist in Sepia officinalis.

I have presented a review and critique of the procedures employed in simultaneous discrimination training experiments using octopuses as subjects. Procedural variables were analyzed statistically for their influence on experimental outcome. The variables most significantly associated with successful discriminations included use of a specific start location for subjects, shock as negative reinforcement, fewer trials per session, more sessions per day, and discriminations based on stimulus brightness. No experiment controlled all potential sources of inadvertent cues, and subjects' performances appeared to be sensitive to exact procedural details. The most common practice diminishing evidence for learning involved reward that coincided with the subject's pre-existing preferences. I found no evidence that sub-optimal experimental designs biased experimental outcomes in any significant and systematic way. Although there is sufficient reason for rejecting results of published simultaneous discrimination training experiments, careful conclusive experiments remain to be performed.

No abstract