Sony DCR-PC1E Sony DCR-PC2E Sony DCR-PC3E Sony DCR-PC4E Sony DCR-PC5E Sony DCR-PC6E Sony DCR-PC7E Sony DCR-PC10E Sony DCR-SC100E Sony DCR-TRV5E Sony DCR-TRV6E Sony DCR-TRV7E Sony DCR-TRV8E Sony DCR-TRV9E Sony DCR-TRV10E Sony DCR-TRV11E Sony DCR-TRV15E Sony DCR-TRV17E Sony DCR-TRV110E Sony DCR-TRV120E Sony DCR-TRV125E Sony DCR-TRV130E Sony DCR-TRV210E Sony DCR-TRV220E Sony DCR-TRV230E Sony DCR-TRV235E Sony DCR-TRV310E Sony DCR-TRV320E Sony DCR-TRV325E Sony DCR-TRV410E Sony DCR-TRV420E Sony DCR-TRV430E
Sony DCR-TRV510E Sony DCR-TRV520E Sony DCR-TRV530E Sony DCR-TRV725E Sony DCR-TRV890E Sony DCR-TRV7000E Sony DCR-TRV7100E Sony DCR-TRV8000E Sony DCR-TRV8100E Sony DCR-VX700E Sony DCR-VX1000E Sony DCR-VX9000E Sony DSR-200P

Panasonic

Panasonic NV-DA1 Panasonic NV-DS1 Panasonic NV-DS5 Panasonic NV-DS8 Panasonic NV-DS11 Panasonic NV-DS12 Panasonic NV-DS15 Panasonic NV-DS25 Panasonic NV-DS27 Panasonic NV-DS33 Panasonic NV-DS35 Panasonic NV-DS37 Panasonic NV-DS55 Panasonic NV-DS77 Panasonic NV-DS88
JVC GR-DV1 JVC GR-DV3 JVC GR-DVF1 JVC GR-DVF10 JVC GR-DVJ70 JVC GR-DVL20 JVC GR-DVL30 JVC GR-DVL40 JVC GR-DVL100 JVC GR-DVL108 JVC GR-DVL140 JVC GR-DVL150 JVC GR-DVL300 JVC GR-DVL308 JVC GR-DVL355 JVC GR-DVL450 JVC GR-DVL555 JVC GR-DVL9000 JVC GR-DVL9200 JVC GR-DVL9500 JVC GR-DVL9700 JVC GR-DVM1 JVC GR-DVM5 JVC GR-DVP1 JVC GR-DVX JVC GR-DVX4 JVC GR-DVX7 JVC GR-DVX8 JVC GR-DVX9 JVC GR-DVX44 JVC GR-DVX77 JVC GY-DV500E

Thomson

Thomson VMD2 Thomson VMD3 Thomson VMD8 VDM8 Thomson VMD10 VDM10 Thomson VMD120 VDM120
Metz CC34 Metz CD34 Metz CE34 Metz CF34 Metz CG34

Grundig

Grundig DLC1 Grunding DLC10 Grundig DLC20 Grundig DLC1000 Grundig DLC1500 Grundig DLC1505
Canon MV 1 Canon MV 3 Canon MV 3MC Canon MV 10 Canon MV 20 Canon MV 30 Canon MV 100 Canon MV 200 Canon MV 300 Canon MV 400 Canon MV 425 Canon MV 430 Canon MV 450
Panasonic NV-DS99 Panasonic NV-DX1 Panasonic NV-DX100 Panasonic NV-EX1 Panasonic NV-EX3 Panasonic NV-EZ10 Panasonic NV-EZ30 Panasonic NV-MX1 Panasonic MV-MX5 Panasonic NV-MX12 Panasonic NV-MX400 Panasonic NV-MX450
Compatibility list with the DV Max version 2.3

enabler

not compatible restrictions compatible
Camcorders With Adaptator
Sony DCR-PC7E. Adaptator VM-LM7
Sony DCR-IP5 Sony DCR-IP7 Sony DCR-PC8E Sony DCR-PC9E Sony DCR-PC100E Sony DCR-PC101E Sony DCR-PC110E Sony DCR-PC120E Sony DCR-TRV20E Sony DCR-TRV25E Sony DCR-TRV30E Sony DCR-TRV50E Sony DCR-TRV240E Sony DCR-TRV330E Sony DCR-TRV620E Sony DCR-TRV720E
Canon MV 3i Canon MV 3iMC Canon MV 4i Canon MV 4iMC Canon MV 5i Canon MV 5iMC Canon MV 20i Canon MV 30i Canon MV 200i Canon MV 300i Canon MV 400i Canon MV 430i Canon MV Canon450i Canon MV 500i Canon MV 1 Canon MV 530i Canon MV 3 Canon MV 550i Canon MV X1i Canon XM-1 Canon XL-1 Canon XL-1S
Panasonic AG-DVC15E Panasonic AG-DVC200 Panasonic AG-EZ35 Panasonic NV-DS28 Panasonic NV-DS30 Panasonic NV-DS38 Panasonic NV-DS89 Panasonic NV-DS110 Panasonic NV-DS150 Panasonic NV-DS990 Panasonic NV-DS999 Panasonic NV-DX110EG Panasonic NV-EX215 Panasonic NV-GS3 Panasonic NV-GS5 Panasonic NV-GX7 Panasonic NV-MX7 Panasonic NV-MX8 Panasonic NV-MX300 Panasonic NV-MX350
JVC GR-DV300 JVC GR-DV2000 JVC GR-DV3000 JVC GR-DVL107 JVC GR-DVL109 JVC GR-DVL157 JVC GR-DVL167 JVC GR-DVL301 JVC GR-DVL307 JVC GR-DVL309 JVC GR-DVL357 JVC GR-DVL367 JVC GR-DVL557 JVC GR-DVL567 JVC GR-DVL767 JVC GR-DVL867 JVC GR-DVL9600E JVC GR-DVL9800E JVC GR-DVLX88 JVC GR-DVP3 JVC GR-DVP7
Canon MV 20.. Adaptator DU-200

Cache protection strategies by western scrub-jays (Aphelocoma californica): hiding food in the shade
Joanna M. Dally1*, Nathan J. Emery2 and Nicola S. Clayton1
1 Department of Experimental Psychology, and 2Sub-Department of Animal Behaviour, University of Cambridge, Cambridge CB2 3EB, UK * Author for correspondence ( jmd54@cam.ac.uk).
Recd 09.01.04; Accptd 23.02.04; Published online 16.06.04
In the presence of conspecif ics, food-caching western scrub-jays (Aphelocoma californica) implement a variety of strategies to reduce the chances of cache theft. This experiment aimed to determine whether the jays could exploit an environmental variable, the level of ambient light, to reduce the transfer of visual information to potential pilferers. Each jay was allowed to cache non-degradable food in two trays, one of which was well lit, whereas the other was in shadow. In some trials the birds cached in private and in others they were observed; however, they always recovered their caches in private. When observed the jays preferentially cached in the shaded tray, whereas both trays were used equally when caching in private. By caching in shaded sites, the quality and transfer of visual information available to the observer may be reduced, thereby making the location of cache sites less certain. These results suggest that western scrub-jays may selectively cache in the shade as a strategy to reduce the chance of cache theft by observing conspecif ics. Keywords: scrub-jay; social cognition; caching; cache protection; corvids
1. INTRODUCTION Food-storing corvids, such as western scrub-jays, hide food for future consumption and rely on memory to recover their caches at a later date. These caches are susceptible to pilfering by other individuals, as a number of species commonly use spatial memory to recover caches that they have seen conspecifics make (Bednekoff & Balda 1996; Clayton et al. 2001; Bugnyar & Kotrschal 2002). Field observations suggest that storers implement a variety of measures to reduce cache theft. These strategies include waiting until conspecifics have left or are distracted before caching (e.g. rooks, Corvus frugilegus; Simmons 1970; Kallander 1978), or returning alone to caches previously hidden in the presence of conspecifics and recaching them in sites unbeknown to the observer (e.g. western scrub-jays; Emery & Clayton 2001; Goodwin 1955; ravens, Corvus corax; Heinrich 1999). Furthermore, ravens will cache behind large objects in the presence of conspecifics. These objects may serve as barriers
Proc. R. Soc. Lond. B (Suppl.) 271, S387S390 (2004) DOI 10.1098/rsbl.2004.0190
concealing information about the location of caches from potential thieves (Bugnyar & Kotrschal 2002). Although field observations are of enormous value in documenting the natural behaviour of these birds, an experimental approach is crucial for understanding the mechanisms underlying these behaviours and determining the effects of experience. Emery & Clayton (2001) tested whether it was the presence of an observer at caching, and the absence of one at recovery, that elicited the storers recovery behaviour. Hand-raised western scrub-jays were allowed to cache either in private or while observed, and then recover their caches in private. Individuals that had prior experience (outside of the experiment) of pilfering another birds caches subsequently re-cached food in new sites, but only when observed during caching. Because the two conditions were identical at the time of recovery, the birds had to remember whether or not they had been watched during caching to know when to re-cache during recovery, and whether to cache into new sites. In the Emery & Clayton (2001) study, as in many of the field observations (e.g. Goodwin 1955), the birds behaviour could be a response to the presence or absence of an observer during caching, coupled with a memory for their presence or absence at recovery. However, recent experiments have shown that when an observer is always present at the time of caching, jays use distance and barriers to reduce the probability of cache theft by conspecifics (Dally et al. 2005). For example, when given the choice of two caching trays, one near to and one far from an observer, the jays cached preferentially in the far tray, but not when caching in private. When given the choice of caching in a tray in view of a conspecific, or one hidden behind a barrier, the jays cached significantly more in the tray that was out of view, again not when caching in private. In both cases, during recovery, caches were recached only from the tray most at risk from pilfering (i.e. the near tray and the in view tray). These results suggest that the jays may have been attempting to minimize the transfer of visual information to the observer during the caching event. In the field some cache sites would also be obscured by light intensity. For example, sites in shady spots would be far less visible than those in bright sunlight. The aim of this experiment is to test whether jays use shade as a cache-protection strategy. If jays act to reduce the visual information available to the observer at the time of caching, they should prefer to cache in shady sites when observed but not necessarily when caching in private. To test this hypothesis, the jays were allowed to cache both in private and when observed. In each condition they were provided with two caching trays that were equidistant from the observers cage. One tray was well lit and the other was placed in the shade. 2. METHODS

Nine sexually mature (four females and five males) hand-raised western scrub-jays were used. The jays had previously participated in a range of experiments ( J. M. Dally, unpublished data), acting as both observers and pilferers. In these studies the caching trays were removed after the caching phase and on some trials the storer subsequently observed another jay pilfering their caches. This experience meant that re-caching behaviour could be expected. Birds were housed in cages measuring 2 m wide by 1 m high by 1 m deep, and were maintained inside on a 12 L : 12 D cycle on a diet of mixed nuts, dog biscuits, various fruits, wax worms and mealworms. The maintenance diet was removed from the home cage the evening before an experimental day, ensuring that the jays were mildly hungry during caching.

2004 The Royal Society

S388 J. M. Dally and others Cache protection strategies

observer

shaded

bright

storer
Figure 1. (a) The experimental set-up with the location of the shady and well-lit sunny cache sites in a storers compartment, and the observers cage. (b) The photograph shows a scrub-jay in one of the storer compartments, which contains two caching trays and a food bowl. A lamp was attached to the side of the compartment at a height of 54 cm above each caching tray. Note that only the lamp above the well-lit (sunny) tray was switched on, resulting in a difference in illumination between the well-lit and shaded trays of 4 : 1 and a difference in temperature of 2 C (see 2). This difference in brightness cannot be detected from this photograph because of the use of flash photography.
The home cages were the site of all caching trials. Each jay was placed in a 1 m 1 m 1 m compartment of the home cage, and these were arranged side by side in rows of three, with an observer jay in the middle compartment and a storer in each of the two outer compartments (see figure 1). There were two caching conditions: in private and observed. During in private trials, the observers view was blocked by solid opaque dividers. Transparent dividers separated the observer from the observed storer during observed trials. The birds received two caching trials, one in each condition, as well as an additional trial in which they played the role of observer. The birds were moved from one compartment to another after each trial to counterbalance both the order in which the birds received these trials, and whether or not they cached in the cage to the left or right of the observer during the observed condition. Both of the storers cages were differentially lit. Opaque metal screens were attached to the top and side of one half of the cage, and transparent sheets to the other. An anglepoise lamp was placed next to the transparent sheet to increase light levels on that side of the cage, and a second lamp was placed by the opaque screen but switched off. This resulted in a shady and a sunny side of the cage. The ratio of illuminance between the sunny and shady sites was 4 : 1. The temperature of the caching substrate was 28 C on the sunny side and 26 C on the shady side. A Sony DCR-PC8E digital handycam recorded caching trials with input from three colour cameras (Sanyo, VCC-6572P). Each camera was positioned ca. 30 cm from the front of each cage. The jays behaviour was recorded throughout the 15 min caching period. During this time the storers were given two caching trays and a 15 cm bowl containing 50 pine nuts. The caching trays (6 cm 25 cm), contained a array of 2.5 cm ice cube moulds filled with corn kibble, with each of the moulds acting as a possible caching site. Each tray was attached to the centre of a wooden board (15 cm 32 cm) and made visuo-spatially unique by attaching an arrangement of Lego bricks along one of the long sides. The trays were positioned in the cage equidistantly from both the observer and the food bowl. One tray was placed on the sunny side of the cage and one on the shady side. At the end of the caching period, the trays and food bowl were removed along with any caches made in the storers cage. The

Proc. R. Soc. Lond. B (Suppl.)
experimenter then recorded the locations of the food items cached in either tray. Opaque dividers were then erected, and a retention period of 3 h followed before the trays were returned to the storers cage, the storers were then able to recover their caches in private. During the recovery period, the experimenter recorded the number of caches recovered, and what proportion of these were eaten or recached. The trays were then removed and the maintenance diet returned to the cage. The videos were scored to verify the number and location of caches and non-parametric statistics (Wilcoxons matched pairs signed-rank test) were used to analyse the data, with condition (in private versus observed), and tray (sun versus shade) as within-subject variables. Alpha was set at 0.05. As Mundry & Fischer (1998) have suggested that inflated p-values can result from the use of modern statistical programs to perform non-parametric statistics on behavioural data with low sample sizes, we performed all of our Wilcoxon tests by hand (Siegel & Castellan 1988).
3. RESULTS (a) Caching The jays cached a median of two items (inter-quartile range (IQR) = 8), and there was no significant difference between the number of items cached on observed (median = 2, IQR = 5) and in private trials (median = 3.5, IQR = 12; n = 9, T = 15, p 0.5). As figure 2 shows, however, there was a significant difference in the number of items cached on the sunny and shady sides of the cage when observed (sun median = 0, IQR = 2; shade median = 5, IQR = 9.25; n = 6, T = 21, p = 0.016), whereas in private no significant differences were found (sun median = 3, IQR = 8.75; shade median = 4, IQR = 13.25; n = 8, T = 21, p = 0.38).
Cache protection strategies J. M. Dally and others S389

number of items cached

0 sun shade observed sun shade in private
Figure 2. A box and whisker plot showing the median number, 25th percentile, and 75th percentile of items cached in the sunny and shady trays for both observed and in private conditions.
(b) Recovery The jays recovered a median of 0% of their caches (IQR = 0.2). The low levels of recovery (three birds in the observed condition and two in the in private condition) meant that it was not possible to detect any differences between the proportion of items recovered after the observed (median = 0, IQR = 0.61) and in private trial (median = 0, IQR = 0.07). There was no significant difference in the proportion of items recovered from either the sunny or shady sides of the cage when observed (sun median = 0.31, IQR = 0.81; shade median = 0, IQR = 0.04; n = 3, T = 4, p = 0.38). The low level of recovery after caching in the in private condition meant a statistical comparison of the proportion of items recovered from the sunny (median = 0, IQR = 0.05) and shady (median = 0, IQR = 0.02) sides of the home cage was not possible. Low levels of recovery also meant that statistical analysis of the proportion of items recovered and then eaten or recached was not possible. Items were eaten from both trays during both observed (sun median = 0.25, IQR = 0.5; shade median = 0.5, IQR = 1) and in private caching conditions (sun median = 0.55, IQR = 0.9; shade median = 0.5, IQR = 1). However, the pattern of re-caching behaviour differed between caching conditions. When birds had cached in private they moved approximately equal amounts from both trays (sun median = 0.45, IQR = 0.9; shade median = 0.5, IQR = 1). By contrast, when they had been observed during caching, birds selectively moved only those items that had been cached in the sunny tray (sun median = 0.75, IQR = 0.5; shade median = 0, IQR = 0).

equally when caching occurred in private. We suggest that this differentiation in site use may stem from the potential threat posed by the observer. By caching only in the shaded tray in the presence of possible thieves, the level and quality of information available for visual transfer, and thus the chance of cache theft, is reduced. One possible explanation for the unusually low recovery rate is that we used pine nuts, which are not as preferred as the wax worms that we have used in previous studies. Nonetheless, the pattern of re-caching was consistent with the prediction, that storers should preferentially re-cache those items to which the observer has had the greatest visual access and so are therefore most at risk of theft. Thus items were re-cached from both trays in the in private condition, but only from the sunny tray in the observed condition. The preference to cache in the shady tray during the observed condition could not be explained in terms of an aversion to the light source, because birds cached equal amounts in both trays during the in private condition. The probability that cache sites were selected based on differential reinforcement can also be discarded as caches were not pilfered, and only a single trial was given in each of the two caching conditions. In contrast to previous studies (e.g. Emery & Clayton 2001), it was not possible for the storer to use the presence or absence of an observer as a cue to caching behaviour. Although a preference for out of view sites has been demonstrated in the presence of an observer, it is not clear whether the birds differentiated between what the observer could see, as opposed to what the storer saw. As such the observer may appear to be absent when the storer is behind the barrier. In this experiment the observer was always in view of the storer. The selective use of the shaded tray when observed during caching therefore raises the intriguing possibility that the jays may act to reduce the transfer of visual information, and therefore use shade as a cache protection strategy. Future studies will address the effect that this cache protection strategy may have on the recovery accuracy of the observer. These results raise the intriguing question of whether the jays understand what an observer can and cannot see. They also add to the other strategies which scrub-jays use to protect their caches such as caching out of view, recaching and caching at a distance from observers (Emery & Clayton 2001; Emery et al. 2004; Dally et al. 2005), which may provide convergent evidence for complex social cognition in scrub-jays.

Acknowledgements This work was supported by grants to N.S.C. from the National Institutes of Health NS35465-05 and MH2602, BBSRC grant S16565, Clare College, and the University of Cambridge. N.J.E. was funded by a Royal Society University Research Fellowship. J.M.D. received financial support from Clare College, the University of Cambridge, the Sir Richard Stapely Educational trust and the Sefton Educational Trust. Thanks go to both Selvino de Kort for his comments on an earlier version of the manuscript, and to reviewer 1 for ideas on future experiments. These experiments comply with the UK Home Office regulations concerning animal research and welfare, as well as the University of Cambridge regulations on the use of animals. Bednekoff, P. & Balda, R. 1996 Social caching and observational spatial memory in pinyon jays. Behaviour 133, 807826. Bugnyar, T. & Kotrschal, K. 2002 Observational spatial learning and the raiding of food caches in ravens, Corvus corax: is it tactical deception? Anim. Behav. 64, 185195.
4. DISCUSSION These results demonstrate that social context and environmental variables affect a jays choice of cache site. Although jays cached preferentially in the shaded tray when observed during caching, both trays were used
S390 J. M. Dally and others Cache protection strategies
Clayton, N. S., Griffiths, D. P., Emery, N. J. & Dickinson, A. 2001 Elements of episodic memory in animals. Phil. Trans. R. Soc. Lond. B 356, 14831491. (DOI 10.1098/rstb.2001.0947.) Dally, J. M., Emery, N. J. & Clayton, N. S. 2005 Cache protection strategies by western scrub-jays (Aphelocoma californica): implications for social cognition. Anim. Behav. (Submitted.) Emery, N. J. & Clayton, N. S. 2001 Effects of experience and social context on prospective caching strategies by scrub jays. Nature 414, 443446. Emery, N. J., Dally, J. & Clayton, N. S. 2004 Western scrub-jays (Aphelocoma californica) use cognitive strategies to protect their caches from thieving conspecifics. Anim. Cogn 7, 3743. Goodwin, D. 1955 Jays and crows recovering hidden food. Br. Birds 48, 181183. Heinrich, B. 1999 Mind of the raven. New York: Harper Collins. Kallander, H. 1978 Hoarding in the rook (Corvus frugilegus). ANSER Suppl. 3, 124128. Mundry, R. & Fischer, J. 1998 Use of statistical programs for nonparametric tests of small samples often lead to incorrect P-values: examples from animal behaviour. Anim. Behav. 56, 256259. Siegel, S. & Castellan Jr, N. J. 1988 Nonparametric statistics for the behavioural sciences. New York: McGraw-Hill. Simmons, K. E. L. 1970 Further observations on food-hiding in the corvidae. Br. Birds 63, 175177.