Modelling the interplay between inadvertent social information use and a pesticide induced foraging bias in bumblebees' crop visitation

This study uses individual-based modeling to demonstrate that while inadvertent social information use and pesticide-induced foraging biases independently alter bumblebee crop visitation patterns without affecting overall colony growth, these behavioral shifts likely modify pesticide exposure risks and should be integrated into risk assessment frameworks.

The Gist

Imagine a bustling city of bumblebees living in a giant hive. These bees are hardworking commuters, flying out every day to find food (nectar and pollen) to bring back to their families. Usually, they are very smart: they know exactly which flowers are full of food, how far away they are, and they try to take the shortest, most efficient route to get the job done.

However, this paper explores what happens when two specific things mess with their usual routine: following the crowd and getting a little "drunk" on pesticides.

Here is the story of the research, broken down into simple concepts:

1. The Setup: The Bee City and the "Oilseed Rape" Fields

The scientist built a virtual world (a computer simulation) to watch these bees.

• The Home Base: A lush, green park in the middle of the map where the bees live. It has lots of different wildflowers.
• The Target: Huge fields of a crop called Oilseed Rape (OSR). These fields are like giant, single-flower supermarkets. They are very close to the hive (the "Close" fields) and some are far away (the "Far" fields).
• The Problem: In the real world, farmers often spray these crops with pesticides. The bees can't see the poison, but it can mess with their brains.

2. The Two "Glitches" in the System

The researcher tested two scenarios to see how they change where the bees go:

Glitch A: The "Copycat" Effect (Inadvertent Social Information)
Imagine you are looking for a good restaurant. You don't know where the best one is, so you see a line of people outside a specific place. You think, "Oh, everyone is going there, it must be good!" and you join the line.

• In the study: Bees see their friends landing on a specific flower patch. Instead of checking if it's actually the best spot, they just follow the crowd.
• The Result: More bees started flocking to the crop fields (especially the ones close to home) just because they saw other bees there.

Glitch B: The "Pesticide Fog" (Foraging Bias)
Imagine you are wearing sunglasses that make everything look the same shade of gray. You can't tell the difference between a delicious apple and a rotten one.

• In the study: When bees are exposed to pesticides (specifically a common one called acetamiprid), their sense of smell and judgment get fuzzy. They can't tell which flowers are "full of food" and which are empty. They start picking flowers randomly, like a drunk person stumbling through a buffet.
• The Result: Because they can't tell the difference, they stop ignoring the crop fields and start visiting them just as often as the wildflowers, even if the wildflowers are better.

3. What Happened in the Simulation?

The researcher ran the simulation thousands of times to see the outcome.

• The Crowd Effect: When bees followed the crowd, they visited the crop fields more often. It was like a viral trend; once a few bees went, the rest followed.
• The Fog Effect: When bees were "foggy" from pesticides, they visited the crop fields even more. They lost their ability to choose the best food, so they ended up at the crop fields by accident.
• The Combination: When both happened at once, the bees visited the crop fields even more, but the two effects didn't multiply each other in a crazy way; they just added up.

4. The Big Surprise: The Hive Didn't Collapse

You might think, "If the bees are following the crowd and getting confused, they must be starving or dying!"

Surprisingly, no.
The total amount of food the colony collected stayed about the same. The number of baby bees born and the number of new queens ready to start their own colonies remained healthy.

• Why? The crop fields were so huge and full of food that even if the bees were making "bad" choices, they still found enough to eat. The crop fields acted as a safety net.

5. The Real Danger: The Invisible Trap

So, if the bees are fine, why does this matter?

The "Trojan Horse" Problem.
The study didn't simulate the bees dying from poison; it simulated them going to the place where the poison is.

• If the bees are following the crowd or are too confused to choose, they spend more time in the pesticide-treated crop fields.
• Even if they don't die immediately, they are getting a constant, low-level dose of poison.
• Think of it like smoking: You might not die from one cigarette, but if you are forced to sit in a smoky room all day because you can't find a way out, you will eventually get sick.

The Bottom Line

This paper tells us that bees aren't just robots following a map. They are social creatures who follow their friends, and they are vulnerable to chemicals that mess up their brains.

• The Good News: The bees are tough; their colonies can survive these changes in behavior.
• The Bad News: Because they are visiting the pesticide fields more often, they are accumulating "sub-lethal" (not immediately fatal) damage. This could make them weaker, less smart, or less able to reproduce in the long run, even if the colony doesn't collapse right away.

The Takeaway: When we assess whether pesticides are safe for bees, we can't just ask, "Does this kill them?" We also have to ask, "Does this make them act weird and go to the dangerous places more often?" Because if they go there more often, the danger is much higher than we thought.

Technical Summary

1. Problem Statement

The study addresses a critical gap in understanding how pollinator behavior influences pesticide exposure in agricultural landscapes. While it is known that pesticides can cause sublethal effects (e.g., impaired cognition, altered foraging motivation), the specific behavioral pathways that increase exposure risk remain poorly understood.

• The Gap: Most existing models assume bumblebees make foraging decisions based solely on resource quality and distance from the colony. They often overlook two key factors:
  1. Inadvertent Social Information (ISI): The use of cues from conspecifics (e.g., seeing other bees foraging) to locate patches without active communication or learning.
  2. Pesticide-Induced Foraging Bias: Evidence suggests that exposure to neonicotinoids (like acetamiprid) impairs the ability of bumblebees to distinguish between rewarding and non-rewarding floral scents, leading to random or suboptimal patch selection.
• Objective: To investigate how the interplay between ISI use and pesticide-induced foraging bias alters patch visitation rates, resource collection, and colony fitness in Bombus terrestris (buff-tailed bumblebees).

2. Methodology

The study utilized an Individual-Based Model (IBM) framework, specifically extending the BEE-STEWARD simulation tool (based on the Bumble-BEEHAVE model).

• Simulated Landscape:
  • A 2D grid (300 × 210 cells) containing one central high-quality grassland patch (nesting habitat with diverse flora) and eight Oilseed Rape (OSR) patches.
  • OSR patches were categorized as "close" (15m from grassland) or "far" (186.9m from grassland).
  • Simulations ran for 50 years to reach population equilibrium, with data analyzed from the 10th year.
• Model Extensions:
  • ISI Use: Implemented as a "copy-the-majority" rule. Foragers prioritize patches where conspecifics are currently active, provided they have already discovered those patches personally. This creates a density-dependent attraction mechanism.
  • Foraging Bias: Modeled based on empirical findings (Kárpáti et al., 2024) regarding acetamiprid exposure. Affected bees lose the ability to assess resource filling levels (profitability) during patch selection. Instead of choosing the most profitable patch, they choose randomly among available flowering patches, effectively ignoring resource quality cues.
• Experimental Design:
  • 200 simulations (50 iterations × 4 scenarios: Control, ISI only, Bias only, ISI + Bias).
  • Metrics: Proportions of nectar and pollen visits to close/far OSR patches, total resource collection (nectar/pollen), adult worker counts, colony numbers, and hibernating queen counts (fitness proxy).
• Statistical Analysis:
  • Generalized Linear Models (GLMs) with beta-binomial error distributions for visitation proportions.
  • Linear Models (LMs) for resource quantities and colony demographics.
  • Effect sizes reported as Odds Ratios (OR) with 95% Confidence Intervals (CI).

3. Key Contributions

• Novel Integration: This is the first study to integrate both inadvertent social information and pesticide-induced cognitive bias into a single spatially explicit IBM for bumblebees.
• Decoupling Mechanisms: It distinguishes between social attraction (ISI) and cognitive impairment (bias) as distinct drivers of foraging patterns, rather than treating them as a single "stress" factor.
• Refined Exposure Risk Assessment: The study shifts the focus from total resource intake to the spatial origin of resources, arguing that increased visitation to treated crop patches (even without total intake changes) elevates exposure risk.

4. Key Results

• Visitation Patterns:
  • ISI Use: Significantly increased nectar visits to both close and far OSR patches. It also increased pollen visits to close OSR patches but had negligible effects on far patches.
  • Foraging Bias: Had a massive impact on pollen foraging, increasing visits to all OSR patches (both close and far) by nearly 3-4 times compared to baseline. It also increased nectar visits to close OSR patches.
  • Interaction: There was no significant interactive effect between ISI and foraging bias. Their effects were largely additive or independent. The presence of one did not amplify the other's impact on visitation rates.
• Resource Collection & Colony Fitness:
  • Despite the pronounced shifts in where bees foraged, the total amount of nectar and pollen collected remained statistically unchanged across scenarios.
  • Colony Demographics: Metrics such as total adult workers, number of colonies, and the number of hibernating queens (reproductive output) were largely unaffected.
  • Note: A slight increase in colony numbers was observed when both factors were present, but this was not considered a major demographic shift.
• Spatial Dynamics: The foraging bias caused bees to distribute their visits more evenly across available patches (reflecting patch availability rather than quality), effectively diverting foragers from the high-quality grassland to the OSR crops.

5. Significance and Implications

• Sublethal Risk Amplification: The study demonstrates that even if pesticide exposure does not immediately reduce colony growth or total food intake, it alters spatial foraging patterns. By driving bees to visit treated crop patches more frequently (due to bias) or attracting more bees to those patches (due to ISI), the probability of individual exposure increases.
• Feedback Loops: Increased visitation to treated patches can lead to higher cumulative doses, potentially triggering a cascade of sublethal effects (immune suppression, reduced drone production, impaired navigation) that may only manifest at the population level over time.
• Risk Assessment Reform: Current pesticide risk assessment frameworks often rely on average exposure estimates. This study argues for incorporating behavioral pathways (social cues and cognitive impairment) into these models to better predict exposure routes.
• Landscape Management: The findings suggest that in landscapes with abundant alternative habitats (like the grassland in the model), the impact of these biases might be buffered. However, in more heterogeneous or resource-scarce landscapes, these behavioral shifts could be more detrimental.

Conclusion:
The paper concludes that inadvertent social information and pesticide-induced foraging biases are critical, previously overlooked behavioral pathways that alter the spatial distribution of bumblebee foraging. While they may not immediately collapse colony populations in resource-rich environments, they significantly increase the risk of pesticide exposure by concentrating foraging activity on treated crops, thereby facilitating the emergence of sublethal effects that threaten long-term pollinator health.