GIANT PACIFIC OCTOPUSBEHAVIOR AND ENRICHMENT

BY MEGAN BRADY, MARK REHLING, JENNI MUELLER AND KRISTEN LUKAS

Introduction

The giant Pacific octopus (Enteroctopus dofleini) is the largest species of octopus in the world and frequently a choice exhibit species in zoos and aquariums (Anderson and Wood, 2001). Octopuses are cephalopods, considered the most intelligent class of invertebrates; they have well developed senses, brains with distinct lobes, and highly manipulative arms (Wood and Wood, 1999). Although individuals live less than five years, their abilities to learn and remember have been observed by both aquarists and divers. Octopuses have long- and short-term memories similar to vertebrates, but separate learning systems for visual and tactile stimuli (Cousteau, 1975). They also have the cognitive capacity to exhibit play behavior similar to that documented in mammals and birds, as is shown by simple and repetitive out-of-context actions using items such as water jets and toys (Mather and Anderson, 1999).

As scientists realized the cognitive capabilities of the octopus, aquarists and zoo keepers began to consider the need for enrichment of captive octopuses. Enrichment is a common husbandry practice used to encourage natural behaviors and stimulate overall physical and psychological well-being (Carlstead, 1996). Intelligent and social species are often candidates for enrichment, as these animals, when housed in non-enriched environments, can often exhibit abnormal sleep patterns or even destructive and unhealthy behaviors that suggest boredom or stress. For non-human animals such as the octopus that live in dynamic environments, enrichment can also help to recreate in captivity some of the complexity of their natural habitat (Anderson and Wood, 2001). Enrichment has most commonly been implemented for mammals, and thus as zoos and aquariums increase the use of enrichment for the octopus it is necessary to evaluate the effects.

There have been abnormal behaviors exhibited by octopuses, especially in a barren laboratory setting, which have implied the need for enrichment. Irregular color patterns, inking, and swimming or jetting into the tank walls are all considered signs of stress in an octopus (Wood and Wood, 1999). Depending on the species, hiding for all hours of the day can also be considered undesirable. Autophagy, or the eating of the appendages, has been seen in laboratory environments but never in an enriched environment (Beigel and Boal, 2006). The use of enrichment offers a way to help the animal cope with stressful situations and can possibly put an end to negative and undesirable behaviors.

There are many different strategies used for enrichment with the octopus. Feeding enrichment activities are easy and effective, and can also improve the animal’s nutrition. Varying between feeding few larger food items and many smaller food items, feeding live instead of frozen food, and presenting unfamiliar prey items are all possible feeding enrichment tactics. Hiding the food or placing it in a puzzle box can also add difficulty and variety to the feeding schedule (Rehling, 2001). Environmental enrichment is typically used to mimic the complex environment of the octopus in the wild and provide opportunities for more natural hunting experiences. Rocks, sand, gravel, and seashells can all provide material for the octopus to build a lair or use for hiding within the exhibit. Pipes and flowerpots have also been found to make good dens for smaller octopus species (Wood and Wood, 1999).

Enrichment seems to provide octopuses with more opportunities to interact with their environment and offers them new places to hide and hunt for prey. Enrichment can also stimulate octopuses to use their natural hunting techniques and to spend more time engaged in feeding behavior (Wood and Wood, 1999). Previous studies have shown that enrichment can increase both the activity level of captive octopuses and visitors’ interest in the exhibit (Rehling, 2001), as well as decreasing undesirable behaviors such as autophagy or jetting into tank walls (Beigel and Boal, 2006).

This study examines enrichment use for captive E. dofleini housed at Cleveland Metroparks Zoo (CMZ), Ohio, U.S.A. One objective of the present study was to determine the effect of enrichment on activity level. A second objective was to determine the type of enrichment that resulted in the octopus exhibiting the highest level of investigation behavior. Enrichment involving food (Hunting Stimuli) was compared to enrichment that did not use food (Interactive Approaches and Environmental Stimuli). We hypothesized that (1) activity level of the octopus would be greater on days when enrichment was present than on days when no enrichment was provided, (2) the octopuses would exhibit higher levels of investigation behavior when a hunting stimulus was used than when interactive or environmental stimuli were present, and (3) the octopuses would use more of the enclosure space when given enrichment.

In correlation with behavior we also examined skin pigment changes. Cephalopods such as the octopus are able to change the coloration and pattern of their skin through the use of chromatophores, small skin cells consisting of elastic sacs surrounded by radial muscles. Contraction of these muscles causes the sacs to expand, increasing the visibility of the pigment granules contained inside and thus altering the skin’s appearance (Messenger, 2001).

Contractions can be of varying strengths, providing a continuum of color alteration using the yellow, red and brown pigments within the sacs (Hanlon et al., 1999). Each chromatophore is attached to a separate nerve ending and can be altered within milliseconds, allowing for changes to happen quickly and with great pattern specificity (Mather and Mather, 2004). Many species also possess layers of iridophores, which produce iridescent blue and green colors via light refraction, and leucophores, which reflect ambient light. These cells are layered under the skin and combine with the chromatophores to give octopuses a wide variety of pigmentation and brightness components (Messenger, 2001). Some cephalopods are known to possess over 30 distinct skin patterns, which can be displayed for various durations in various permutations (Hanlon et al., 1999). This complexity makes understanding octopus color changes difficult, and little is known of their detailed relationship with behavior. Our objective was to examine how the skin pigmentation of the octopus varied with and without enrichment.

The final objective of this study was to determine the effect of visitor presence on the reactivity of the octopus. ‘Reactive behaviors’ were defined as actions intended to put space between the octopus and a threat, such as jetting backwards or shrinking. We predicted that the presence of visitors within 0.3 m of the exhibit would cause the octopus to show greater reactivity than when no visitors were present.

Materials and methods

For this study, one male and one female octopus were used while sequentially exhibited at Cleveland Metroparks Zoo in the Primate, Cat, and Aquatics building. The octopuses were housed in a tank of approximately 1,268 liters (that is remote from a larger sump). The exhibit was approximately 1.22 m3 with viewing from two adjacent sides and a fiberglass rock wall forming the opposite sides. Along the back wall was a crevice which the animals could use as a den. The octopus was typically fed a single prey item every other day using a food pole; if actively exploring or hunting, additional food was offered.

Three types of enrichment were used: Hunting Stimuli, Interactive Approaches, and Environmental Stimuli. Hunting Stimuli were various prey puzzles which replaced the usual pole feeding and acted as the sole source of food for that day; these provided both environmental interaction and the opportunity to perform cognitive and hunting behaviors. Environmental Stimuli included naturalistic objects temporarily added to the enclosure, such as unfamiliar plants, rocks, or seashells; these objects were used to mimic the diverse environment of the octopus and promote natural behaviors. Interactive Stimuli involved physical contact or non-food objects; these promoted tactile and investigative behaviors.

Behavioral data were collected from October 2004 to June 2006 using one-hour continuous focal observations, using Noldus Observer® 3.0 software and the Psion Workabout© handheld device. Presence and type of enrichment were recorded at the start of each observation. Dependent variables recorded continuously included subject behavior, location within exhibit, visitor presence, and skin color. The ethogram for each variable was exhaustive and mutually exclusive (see Appendix 1). A total of 92 hours of data were collected, including 25 hours of observation on the male octopus and 67 on the female. Behavioral activity budgets and exhibit location are based upon percent of total time. Data comparing specific behaviors, enrichment use, and color between enrichment conditions are given as average time in minutes/hour; standard deviation bars are shown.

Results

When enrichment was absent the octopuses most often were resting, followed by locomotion (Fig. 1). With the presence of enrichment resting decreased by two-thirds and locomotion by nearly three-fourths. The octopuses increased object contact and other behaviors when enrichment was present. Active and passive investigation were not displayed by the octopuses when enrichment was absent but combined to nearly half the activity budget when enrichment was present.

The mean occurrences of resting were not significantly different between the male and female when compared both with and without enrichment (Fig. 2). Locomotion in the absence of enrichment was also not significantly different. Figure 2 shows that without enrichment the octopuses each rested an average of 43 ± 0.74 (SEM) min/hour, and with enrichment this decreased to 13.9 ± 3.61 min. The octopuses also significantly decreased locomotion in the presence of enrichment, though the male locomoted at a greater rate than the female.

Enrichment type affected levels of active investigation (Fig. 3). When presented with hunting devices the octopuses averaged 27.41 ± 12.44 (SEM) min. of investigation. Environmental stimuli elicited little activity in either octopus.

Figure 1. Activity budgets of octopuses with and without enrichment. No significant differences between subjects were seen and data were combined; includes 18 hours of data without enrichment and 74 hours with enrichment. The octopuses decreased resting and locomotion when enrichment was present. No active investigation was seen in the absence of enrichment. The presence of enrichment resulted in more activity and an increase in behavioral diversity in the octopuses.

No enrichment

Enrichment

Figure 2. Effect of enrichment on average time spent resting and locomoting. ‘E’ signifies days with any form of enrichment; ‘No E’ signifies days without enrichment. No significant differences between subjects were seen in ‘Rest – No E’, ‘Rest – E’, or ‘Locomote – No E’, and data were combined. The octopuses significantly decreased resting in the presence of enrichment. The female significantly decreased locomotion when enrichment was present; the male also decreased the behavior, though not significantly. The decrease in these behaviors when enrichment was present was likely due to the increase in investigation and other active behaviors.

male responded most to interactive stimuli, averaging 30 ± 2.15 (SD) min/hour, while the female averaged 1.06 ± 1.84 (SD) min. of active investigation toward these devices.

Few abnormal behaviors were seen during this study and all those recorded were displayed by the female. These instances occurred on five separate days over the course of 15 months and included short bouts of both squirting and head-bobbing. The behaviors occurred on both enriched and non-enriched days, and in the presence and absence of visitors.

When the effect of enrichment on exhibit use was compared for the male and female, many individual differences were apparent. The female octopus spent over 70% of her time in the den when no enrichment was present. When given enrichment her use of the exhibit was more evenly distributed and she remained in the den less than 20% of the time. With enrichment the female was most often in the center of the exhibit, an area highly viewable to the public. Unlike the female, the male did not enter the den when there were no enrichment devices in the exhibit, but instead favored the right side of the enclosure (83%). In the presence of enrichment he did utilize all areas of the exhibit, but less evenly than the female due to his tendency to remain in the right-side portion. These results are shown in Fig. 4.

The octopuses displayed pale coloration significantly more often without enrichment and significantly more than the other colors during this condition. When enrichment was present the octopuses most often displayed dark mottled coloration, and the occurrence of this color was significantly greater than without enrichment. The octopuses showed no significant differences in the display of light mottled and flush colorations.

Lastly, the effect of visitor presence on the behavior of the octopuses was evaluated. ‘Present’ was defined as being within 0.3 m of the exhibit glass, while ‘Absent’ was defined as more than 3 m from glass. To examine this, data from both octopuses were combined and a subset was analyzed consisting of (a) observations when visitors were present for more than 70% of the total time (18 total observations), and (b) observations when visitors were absent over 70% of the total time (29 total observations). These data therefore do not include observations with inconsistent levels of visitor presence. No significant differences were seen in behavior, exhibit location, or color, and no reactive behaviors were seen in the presence of visitors.

Figure 5. Changes in skin coloration in relation to enrichment presence.No significant differences between subjects were seen and data were combined. Includes 74 hours of data with enrichment and 18 without enrichment. Light mottled and flush colors showed little difference between conditions. Pale was seen significantly more often without enrichment, while dark mottled was seen significantly more with enrichment.

Discussion

Our results supported the first hypothesis, showing that the giant Pacific octopuses did exhibit more active behaviors when enrichment was present than when no enrichment device was provided. When an enrichment device was present in the exhibit, the animals spent less time resting and more time engaged in active investigation of the enrichment device (Figs. 1 and 2). A study by Beigel and Boal (2006) reported similar findings, concluding that the presence of environmental enrichment resulted in the octopus increasing time spent manipulating tank objects. This suggests that enrichment devices encourage exploratory behavior. The presence of enrichment also significantly decreased resting behavior. Wild giant Pacific octopuses may be active and away from the den site nearly half the time (Mather et al., 1985), and the results indicate that enrichment may provide a method to replicate this natural behavior in a captive setting. The results also show that the octopuses displayed a greater variety of behaviors in the presence of enrichment. Beigel (2004) also found an increased behavioral diversity in the octopus through habitat enrichment, as well as a decrease in destructive behaviors.

The addition of new items to the tank appears to stimulate the exhibit animal to engage in more natural behaviors and as a result spend less time resting and more time visible to the public. Anderson and Wood (2001) found similar results when octopuses were provided prey puzzles and enrichment such as live food or fish fluids; the octopuses remained active for hours at a time without showing signs of habituation or learning. Wood and Wood (1999) found that after enrichment was provided to an octopus for an extended period of time the animal’s activity levels increased and the amount of time spent in the lair decreased. Although the octopuses in this study decreased locomotor behavior when given enrichment, their total activity was greater due to high levels of investigatory behavior. The decrease in locomotion was likely due to the octopuses remaining in one part of the exhibit when investigating enrichment devices (Wood and Wood 1999).

The second hypothesis, that the octopuses would exhibit more active investigation in the presence of Hunting Stimuli than in the presence of the other two types of enrichment, was supported by the data for the female octopus but not entirely by the data for the male (Fig. 3). The female showed much higher levels of active investigation in the presence of the hunting enrichment, but the male showed near equal levels of active investigation for both Hunting Stimuli and Interactive Approaches, and very little active investigation for the Environmental Stimuli. Like many vertebrate predators the giant Pacific octopus has an arrhythmic activity pattern based largely on the availability of food (Mather et al., 1985). Enrichment devices that replicate hunting stimuli are therefore likely to engage the octopuses and promote higher activity levels. However, the data suggest that other forms of enrichment may engage different subjects to varying degrees depending on individual attributes. While other studies have shown an increase in active investigation with the provision of habitat enrichment (e.g. Beigel and Boal, 2006), no other studies have compared different types of enrichment for the octopus, and further work in this area will help elucidate individual preferences.

Abnormal behaviors such as stereotypies are common in captive animals and are often associated with stress or boredom (Anderson and Wood, 2001; Carlstead, 1996). A primary goal of enrichment is to decrease the development and occurrence of these by increasing an animal’s mental stimulation and performance of natural behaviors. Octopuses are known to exhibit abnormal behaviors in some captive settings (Wood and Wood, 1999), and stereotypies were displayed by the female at CMZ. The occurrences were rare and had no apparent similarities: behaviors were shown both with and without enrichment (Hunting Stimuli), in the presence and absence of visitors, and during morning, afternoon, and late afternoon hours. Given the infrequent incidence of these behaviors it is unlikely that the performance is habitual, but instead it may have been a reaction to a temporary environmental variable such as ambient noise or keeper presence. Further study of reactions of individual octopuses to the environment both inside and outside of their exhibit tank may help to clarify the motivation behind this type of abnormal behavior performance.

Enclosure usage of the male varied little between enriched and non-enriched days, with the only apparent differences being increased den usage and less time in the right-side portion on enriched days. He did not enter the den on non-enriched days, instead most often resting on the right side. The difference in den use may be due to him taking Hunting and Environmental Stimuli into the den, which allowed him to eat and/or interact with the items out of view of visitors. It is not known why the male seemed to favor the right side of the enclosure when he was visible. Wild octopuses maintain dens within their home ranges primarily for protection from predators and regularly change the exact sites (Mather et al., 1985). It is possible that the male octopus did not feel threatened within the enclosure and therefore chose not to rest within the den provided, or that during the period of data collection the right corner of the exhibit was his ‘den’. Further studies could examine enclosures in more detail to determine if octopuses favor particular aspects of their enclosures such as plants, rocks, or substrates; this in turn may cause them to spend large amounts of their time within that portion of the exhibit. Unlike the male, the female showed more even spatial distribution in the presence of enrichment and thus supported the third hypothesis (Fig. 4). On days when no enrichment device was provided she averaged over 60% of her time in the den, while enriched days saw decreased den use and increased visibility in all other quadrants. A study by Peters and Powell (2005) had similar results, finding that during an enriched state the octopus would increase exhibit use, moving more throughout the bottom and middle areas of the tank. They also found that by adding a food element to the enrichment the octopus would use more areas of the tank, especially the top and front portions of the exhibit. This suggests that even when the octopuses are not directly in contact with the enrichment devices, they are more active when enrichment is present in the exhibit.

Expanding or contracting chromatophores is a neural event based primarily upon visual input (Hanlon et al., 1999). This color-changing ability of cephalopods has long been considered an adaptation for camouflage, both to escape predators and to approach prey. More recently it has been shown to be a method of both intra- and interspecific communication: for example, during mating interactions of Loligo pealei (Hanlon et al., 1999), and toward prey during foraging of Octopus cyanea (Mather and Mather, 2004). Modern theories also suggest that signal gradation may reflect neuronal responses to emotions such as stress, fear, or aggression (Messenger, 2001). The octopuses in this study did show differences in coloration depending on enrichment presence, with pale skin more prevalent without enrichment and dark mottled pigmentation more common with enrichment (Fig. 5). Pale, light mottled, and dark mottled colorations are common body patterns used by cuttlefish to blend with the stones and sand grains within the environment (Messenger, 2001), and it is possible that these color changes similarly reflected the chosen environment of each octopus. Without enrichment both octopuses spent more time resting (see Fig. 1) and the female spent much more time within the den (Fig. 4); given the lack of movement and changing environments, the need for cryptic coloration is negated and hence the individuals were more often pale. Conversely, mottled patterns are commonly seen in cephalopods when the environment is irregular or changing (Messenger, 2001). When enrichment was present and the octopuses used more of the exhibit and spent more time in active behaviors, the dark mottled pattern may have been a reaction to the continuous environmental changes due to their movement. The results are therefore not conclusive as to whether the enrichment itself stimulated color change in the octopus and, if so, how this reflects the neuronal effects of enrichment. Current understanding of skin pigmentation in relation to behavior is limited to only prominent motivations, such as that seen during mating and aggressive interactions (Messenger, 1999). Future work is needed to examine color change in relation to solitary behaviors such as investigation.

Our final hypothesis, that the presence of visitors close to the exhibit would result in the octopuses exhibiting more reactive behaviors, was not supported by the results. Visitor presence did not affect the animals’ overall behavior nor result in reactive behaviors; additionally, the octopuses did not spend more time out of view of the public and did not increase their den usage in the presence of visitors. A study by Mather and Anderson (1993) found that individual octopuses often prefer one of three principal behaviors when responding to a threat: activity, reactivity, and avoidance. ‘Activity’ was defined by an octopus approaching a threat, ‘reactivity’ as the octopus putting distance between himself and a threat, and ‘avoidance’ as the octopus remaining in the den or hiding during feeding or threat tests (Mather and Anderson, 1993). It is possible that the two octopuses observed in this study could be grouped in the ‘active’ category of behavior variation and that the presence of visitors therefore did not cause them to hide or display reactive behaviors.

Our findings suggest that the use of enrichment for the giant Pacific octopus could be beneficial as it encourages the animal to display a wider range of natural behaviors, be more active, and use a greater percentage of the enclosure space, thereby increasing visibility to the public. Increased behavioral diversity might also draw visitors’ attention to the exhibit and provide greater educational opportunities in a zoo or aquarium setting.

Acknowledgements

References

Megan Brady ¹, ², Mark Rehling ¹, Jenni Mueller ¹, ², and Kristen Lukas ¹, ²;
¹ Conservation and Science, Cleveland Metroparks Zoo, Cleveland, OH 44109,
² Department of Biology, Case Western Reserve University, Cleveland, OH 44106.
Requests for reprints should be sent to Kristen Lukas at kel@clevelandmetroparks.com.

Appendix 1. Ethogram