Agitated honeybees exhibit pessimistic cognitive biases

Abstract

Whether animals experience human-like emotions is controversial and of immense societal concern [1-3]. Because animals cannot provide subjective reports of how they feel, emotional state can only be inferred using physiological, cognitive, and behavioral measures [4-8]. In humans, negative feelings are reliably correlated with pessimistic cognitive biases, defined as the increased expectation of bad outcomes [9-11]. Recently, mammals [12-16] and birds [17-20] with poor welfare have also been found to display pessimistic-like decision making, but cognitive biases have not thus far been explored in invertebrates. Here, we ask whether honeybees display a pessimistic cognitive bias when they are subjected to an anxiety-like state induced by vigorous shaking designed to simulate a predatory attack. We show for the first time that agitated bees are more likely to classify ambiguous stimuli as predicting punishment. Shaken bees also have lower levels of hemolymph dopamine, octopamine, and serotonin. In demonstrating state-dependent modulation of categorization in bees, and thereby a cognitive component of emotion, we show that the bees' response to a negatively valenced event has more in common with that of vertebrates than previously thought. This finding reinforces the use of cognitive bias as a measure of negative emotional states across species and suggests that honeybees could be regarded as exhibiting emotions.

Figure 1

Protocol for Cognitive Bias Experiment with Olfactory Conditioning of Honeybees Honeybees were trained for six trials with each stimulus (CS) in a pseudorandomized sequence. The CS+ odor was a ratio of 1 part 1-hexanol to 9 parts 2-octanone; the CS− was a 9:1 ratio of the same two odors. After conditioning, bees were placed either in a group that was exposed to 60 s of shaking or in a control group. All bees began the testing session within 300 s of the manipulation. They were tested with each CS and three novel, intermediate ratios of the same two odors. All test trials were unreinforced, and the order of test odors was randomized across subjects.

Figure 2

Vigorous Shaking for 60 Seconds on a Vortecizer Reduced the Levels of Biogenic Monoamines in Honeybee Hemolymph around 300 Seconds Later Dopamine (DA, F_1,24 = 7.79, p < 0.011), octopamine (OA, F_(1,24) = 5.16, p < 0.034), and serotonin (5HT, F_1,24 = 8.84, p < 0.007) all decreased, but the level of tyramine (TA, F_1,24 = 0.041, p = 0.841) did not change. n_control = 12, n_shaken = 13. Actual (untransformed) mean values are as follows: DA, $\overline{x}$ unstressed = 8.42 ± 3.05 μM, stressed = 4.30 ± 1.75 μM; OA, $\overline{x}$ unstressed = 87.9 ± 25.3 nM, stressed = 63.4 ± 24.2 nM; 5HT, $\overline{x}$ unstressed = 2.38 + 1.02 nM, stressed = 0.781 ± 0.522 nM; TA, $\overline{x}$ unstressed = 14.4 ± 4.58 nM, stressed = 12.8 ± 4.07 nM. Error bars represent ± standard error of the mean (SEM).

Figure 3

Shaken Honeybees Exhibit a Pessimistic Cognitive Bias When honeybees were subjected to shaking and then tested with the CS+, the CS−, and three novel odors, the slope of the gradient of the line indicating the proportion of bees that extended their proboscis [i.e., P(response)] became steeper (shaking × test odor interaction: χ₁² = 8.08, p = 0.005). The bees were significantly less likely to respond to the CS− and its adjacent novel odor (^∗p < 0.05). The data represent the pooled responses from all three unconditioned stimulus (US) treatments reported in Figure S1. n_control = 69, n_shaken = 78; error bars represent ± 1 SEM.