Temporal consistency of judgement biases in bumblebees

This study demonstrates that while bumblebees' response latencies in judgment bias tasks reflect stable individual traits, their actual chamber choices are largely state-dependent and influenced by transient external factors.

The Gist

Imagine you are trying to figure out if a friend is generally a "glass-half-full" optimist or a "glass-half-empty" pessimist. You might ask them, "If you see a cloudy sky, do you think it's going to rain (bad) or just a little drizzle (okay)?"

Scientists have been using similar tests on animals to see if they have an "emotional mood." But here's the big question: Is that mood a permanent part of their personality (a trait), or does it just change depending on whether they had a good breakfast or a bad day (a state)?

This paper is like a detective story about bumblebees trying to solve that mystery.

The Setup: The Bee "Mood Ring"

The researchers set up a game for bumblebees. Think of it like a video game with two doors:

• The Blue Door: Always leads to a giant, delicious sugar treat (High Reward).
• The Green Door: Always leads to a tiny, sad drop of weak sugar water (Low Reward).

The bees quickly learned which door was which. Once they were pros at this, the scientists introduced the "test." They showed the bees colors that were in between Blue and Green—like a murky teal or a muddy green. These were the "ambiguous" clues.

• If a bee chose the Blue side when seeing a murky color, it was acting optimistic (thinking, "Hey, maybe this is the big treat!").
• If it chose the Green side, it was acting pessimistic (thinking, "Nah, this is probably the weak stuff").

The scientists tested the bees on three separate days to see if their "mood" stayed the same or if it was just a fleeting feeling.

The Two Clues: Speed vs. Choice

The researchers looked at two different things the bees did:

1. How fast they made a decision (Latency).
2. Which door they picked (Optimistic vs. Pessimistic).

1. The Speed: The "Personality" Clue

The bees were surprisingly consistent in how fast they made their choices. Some bees were just naturally "speed demons," zooming toward the doors no matter what color they saw. Others were "slow and steady," taking their time to think.

• The Metaphor: Imagine a group of runners. Some always sprint, and some always jog. Even if the race changes, their running style stays the same.
• The Finding: This speed was a stable trait. It was part of the bee's personality.

2. The Choice: The "Mood" Clue

However, when it came to which door they picked for the tricky, in-between colors, the bees were all over the place. One day, a specific bee might be super optimistic and pick the Blue door. The next day, the same bee might be pessimistic and pick the Green door.

• The Metaphor: Imagine you are at a buffet. Sometimes you feel brave and grab the fancy dessert (optimistic). Other times, you feel cautious and stick to the salad (pessimistic). But you aren't a "dessert person" or a "salad person" by nature; your choice just depends on how you feel right now.
• The Finding: The bees' choices were transient states. They weren't showing a permanent personality trait of being "optimistic bees." Instead, their choices were likely influenced by tiny, momentary factors like how hungry they were, the weather, or just how they felt that morning.

The Big Takeaway

Before this study, scientists worried that when they tested animals for "mood," the animals might just be learning the rules of the game or getting confused by repeated testing.

This paper found two cool things:

1. The "Active Choice" Game Works: By using a game where the bees had to pick a door (rather than just ignoring a signal), the scientists avoided confusing the bees. The bees didn't get bored or learn that the "mystery colors" meant "no reward." They stayed genuinely unsure, which is exactly what you need to test a mood.
2. Mood vs. Personality: In bumblebees, how fast they think is their personality, but what they decide is their mood.

Why Does This Matter?

Think of it like checking the weather. If you want to know if it's going to rain (a temporary state), you look at the clouds. You don't look at the type of soil (a permanent trait).

This study tells us that when we see a bee (or any animal) making a "pessimistic" choice, it's likely because they are having a "bad day" or a temporary shift in mood, not because they are a naturally grumpy bee. This helps scientists understand animal emotions better and ensures that when we try to improve animal welfare, we are addressing their temporary feelings, not just their unchangeable nature.

In short: Bees have consistent personalities (speed), but their optimism is just a fleeting mood (choice). And the best way to catch that mood is to let them make a real choice, not just sit and wait.

Technical Summary

1. Problem Statement

Judgment bias tasks are widely used to infer affective states (emotions) in animals by assessing how they interpret ambiguous stimuli. However, a critical unresolved question is whether these responses reflect transient affective states (mood) or stable individual traits (personality).

• The Gap: While previous studies have shown that animals can be manipulated into "optimistic" or "pessimistic" states, it is unclear if baseline judgment biases are consistent over time without manipulation.
• Methodological Challenge: Many existing studies use "go/no-go" paradigms, which are susceptible to confounding factors like learning (animals learning ambiguous cues are unrewarded) and motivational changes. There is a need to determine if active choice paradigms can better disentangle stable traits from transient states in invertebrates.

2. Methodology

The study utilized bumblebees (Bombus terrestris) to investigate the repeatability of judgment biases across three consecutive testing sessions without any experimental manipulation of their affective state.

• Subjects: Female worker bumblebees, tagged within 24 hours of eclosion to control for age.
• Experimental Design:
  • Training: Bees were trained to associate two distinct colors (Blue and Green) with different reward magnitudes: a High Reward (50% sucrose) and a Low Reward (30% sucrose). The location of rewards was counterbalanced.
  • Testing (Active Choice Paradigm): Bees were presented with five visual cues: the two trained colors and three ambiguous intermediate colors (Near-High, Medium, Near-Low).
  • Procedure: Unlike go/no-go tasks, bees had to actively choose between two chambers. During testing, both chambers contained distilled water (0 reward), ensuring the choice relied solely on the bee's interpretation of the cue.
  • Repetition: The full testing protocol was repeated on three consecutive days. Refresher trials with rewarded cues were interspersed to maintain learning without resolving the ambiguity of the test cues.
• Data Collection:
  • Choice: Whether the bee entered the chamber associated with the High Reward (Optimistic) or Low Reward (Pessimistic).
  • Latency: The time taken to enter a chamber.
• Statistical Analysis:
  • Linear Mixed-Effects Models (LMM) and Generalized Linear Mixed-Effects Models (GLMM) were used to analyze latency and choice probability, respectively.
  • Repeatability ($R$): Calculated using the rptR package to quantify the proportion of variance explained by individual identity (stable traits) versus residual variance (state-dependent factors).

3. Key Contributions

• Paradigm Validation: The study demonstrates that active choice tasks effectively prevent the "loss of ambiguity" often seen in go/no-go tasks. Bees maintained graded responses to ambiguous cues across three days, indicating they did not learn that the cues were unrewarded.
• Trait vs. State Differentiation: The research provides empirical evidence distinguishing between behavioral components that are stable traits (latency) and those that are transient states (choice direction).
• Baseline Variability: It establishes a baseline for natural variability in bumblebee judgment, showing that in the absence of manipulation, bees exhibit negligible individual bias consistency in their choices.

4. Results

• Learning and Discrimination: Bees successfully learned the color-reward associations, showing significantly faster latencies for high-reward cues compared to low-reward cues during training.
• Response to Ambiguity: During testing, bees showed a graded response:
  • They chose the high-reward chamber more often for cues closer to the positive color.
  • Latency increased as cues became more ambiguous or closer to the negative color.
• Repeatability of Latency (Stable Trait):
  • Choice latency showed moderate and significant repeatability ($R = 0.34$, 95% CI: 0.20–0.49).
  • This indicates that individual bees consistently differed in their speed of decision-making, suggesting a stable, trait-like underlying component (potentially linked to personality or metabolic constraints).
• Repeatability of Choice (Transient State):
  • The probability of making an "optimistic" choice (selecting the high-reward chamber) showed negligible repeatability ($R \approx 0$).
  • Even when excluding conditioned cues and analyzing only ambiguous cues, individual consistency remained non-significant.
  • This suggests that the direction of the judgment bias is highly state-dependent and influenced by situational factors rather than being a fixed individual trait.

5. Significance and Conclusion

• Interpretation of Judgment Bias: The findings suggest that in the absence of experimental manipulation, judgment biases in bumblebees are not stable traits but rather reflect transient affective states. Consequently, when manipulations (e.g., stress or positive reinforcement) do shift these choices, the shift is likely a genuine change in affective state rather than a revelation of a pre-existing personality trait.
• Methodological Recommendation: The study advocates for the use of active choice tasks over go/no-go paradigms for invertebrates. Active choice tasks successfully isolate transient affective states from stable behavioral traits by minimizing learning confounds and allowing for the measurement of both decision speed (trait) and decision direction (state).
• Ecological Implication: The dissociation between stable decision speed and flexible decision direction allows bees to maintain consistent activity levels while dynamically adjusting their risk assessment based on current environmental conditions.

In summary, this paper refines the methodology for assessing animal emotions, demonstrating that while the speed of judgment in bumblebees is a stable trait, the optimism/pessimism of the judgment itself is a fluid, state-dependent response.