Foraging as sampling without replacement: A Bayesian statistical model for estimating biases in target selection.

Foraging entails finding multiple targets sequentially. In humans and other animals, a key observation has been a tendency to forage in 'runs' of the same target type. This tendency is context-sensitive, and in humans, it is strongest when the targets are difficult to distinguish from the distractors. Many important questions have yet to be addressed about this and other tendencies in human foraging, and a key limitation is a lack of precise measures of foraging behaviour. The standard measures tend to be run statistics, such as the maximum run length and the number of runs. But these measures are not only interdependent, they are also constrained by the number and distribution of targets, making it difficult to make inferences about the effects of these aspects of the environment on foraging. Moreover, run statistics are underspecified about the underlying cognitive processes determining foraging behaviour. We present an alternative approach: modelling foraging as a procedure of generative sampling without replacement, implemented in a Bayesian multilevel model. This allows us to break behaviour down into a number of biases that influence target selection, such as the proximity of targets and a bias for selecting targets in runs, in a way that is not dependent on the number of targets present. Our method thereby facilitates direct comparison of specific foraging tendencies between search environments that differ in theoretically important dimensions. We demonstrate the use of our model with simulation examples and re-analysis of existing data. We believe our model will provide deeper insights into visual foraging and provide a foundation for further modelling work in this area.

Introduction

Foraging typically refers to the process of searching for and gathering dispersed food. As would be expected given the survival value of this skill, a well-studied question in animal behaviour is the extent to which foraging can be thought of as efficient (e.g. [1, 2]). That is, does the forager’s path and food source selection maximize energy intake and minimize energy expenditure and risk? For contemporary humans, most foraging for food takes place in supermarkets, but foraging can be more broadly defined as search for multiple instances of different categories of targets. Many tasks involve this more general definition of foraging (e.g. a security officer monitoring a crowded event; a radiologist searching for all signs of cancer in an X-ray [3]).

The human foraging literature has been directly inspired by animal behaviour research, including research on how humans find patches [4] and how they terminate their search and move onto another patch [3, 5–8]. In the latter context, an influential modelling framework has been marginal value theorem (MVT) [9], which predicts that an optimal forager should leave a patch when the “instantaneous rate of return” from a given patch drops below the average rate of return across all the patches (including travel time, during which no elements can be collected). Recent work has shown that human observers can forage optimally [3], although interestingly this may not hold in all cases e.g. older adults appear to show non-optimal behaviour, staying too long in each patch to adhere to MVT [8], and other studies have also found deviations from optimality [10]. Thus, when considering search termination, there are good theoretical models of behaviour that can be used to understand the cognitive processes underlying performance.

Another well-established feature of foraging is that animals tend to search for food in ‘runs’ of one particular food type, particularly when prey are cryptic [11, 12]. (See Figs 1 and 2 for examples.) Pollinators exhibit a tendency known as ‘flower constancy’, where they selectively visit flowers of a particular category and ignore equivalently rewarding flowers from other categories. This tendency seems to be context-sensitive in many species, suggesting it is a behavioural adaptation e.g. [13]. These observations led to research into the concept of the ‘search image’, where animals direct selective attention to particular features of prey in order to facilitate fast detection [14, 15]. A similar concept of ‘attentional templates’ (the representation of the search goal in working memory) has also been proposed for human visual search [16]. Target switching is important because it presents a promising model task for understanding the coordination of cognitive and physical effort. Selecting targets based on proximity, that is, selecting all the nearby targets before moving to a new area, minimizes the distance of travel, but the cost is the high cognitive load associated with switching between multiple target templates in working memory. Selecting all the targets of one category before moving on to the next category will minimize the cognitive load, at the cost of increased distance of travel between targets. Which of these two strategies is more efficient will depend on features of both the environment (the distance between targets, the number and complexity of the different types of targets) and the individual (their physical as well as cognitive capabilities). Whether and how humans and other species adjust their strategies in a way that is sensitive to these constraints is a question with broad implications for understanding the efficiency of behaviour more generally.

Fig 1

Example data from Dawkins (1971).

The first 100 grains taken by one chick in experiment 1.

Fig 2

An example of how foraging tasks can be thought of as spatial, using example data from Kristjánsson et al. (2014).

The ‘path’ taken by the forager is indicated by the white line, with the numbers indicating the order of targets taken (first three selections only).

When focusing on target switching, behaviour similar to that seen in nonhuman animal studies has now been replicated multiple times in human adult participants [17-24]: in a task where participants need to collect as many targets as possible (while ignoring distractors), they tend to switch target type regularly when the task is relatively easy (e.g. when the targets can be distinguished from the distractors by a single feature dimension, such as colour, in an analogous manner to feature search in classical visual search paradigms). However, when participants are completing a more difficult search task (e.g. a conjunction search, where the target can only be distinguished from the distractor using two or more feature dimensions) they tend to search in ‘runs’, exhaustively searching for one target type before switching to another.

Human foraging studies have also highlighted that there can be surprising heterogeneity between participants in their search strategies. Some participants have been termed ‘superforagers’, showing no difference in foraging strategy for feature and conjunction type searches [17, 18, 25], potentially implying that they are able to hold multiple target templates simultaneously in working memory. However, recent work has also suggested that this group of people may be better described as employing a different strategy; working memory performance is not a good predictor of these individual differences [26], and in fact, it may not be a stable trait. Instead, people may be able to flexibly adjust their strategy in accordance with task demands [25, 27, 28]. It has even been suggested that ‘superforagers’ is a misnomer, as these participants may show more errors than ‘normal’ foragers, and instead these strategies may be sub-optimal responses to the task demands [29]. Interestingly, children under the age of 12 find conjunction foraging very difficult, indicating the important role of executive functions in this task, and highlighting that our understanding of the cognitive processes underlying run behaviour remains incomplete [24].

Problems with run statistics

An episode of foraging entails a sequence of target-detection responses which have both a spatial and a target identity dimension. With each target selection response in the sequence, the distribution of remaining targets changes on both of these dimensions. In this respect, foraging data presents a unique analytical challenge relative to single-target search.

To study target switching, many previous studies have used a set of similar, highly correlated dependent measures: the target switch rate, the average number of runs per trial, or the average run length per trial. An issue with all of these measures is that the ground truth of the search area will influence target switching; the relative proximity and number of targets of the other category makes switches more likely. This is problematic for comparing run statistics across different conditions or studies that vary the distributions of target locations and categories. More fundamentally, run statistics do not adequately represent the underlying cognitive processes that might be determining foraging behaviour. Ideally, we would to model behaviour with respect to the relative contribution of parameters such as the tendency towards target constancy or switching target type, in conjunction with possible biases for one target type over another.

Run statistics also leave aside the intrinsic spatial aspects of foraging tasks. The effect of the distribution of foraging targets relative to the forager is central to understanding foraging efficiency: for example, patch leaving is affected by distance between patches [3], and heavy bees have a stronger proximity bias than lighter ones [30]. Despite this, relatively little work has looked at target switching in direct relation to the effect of target proximity (although see [31] for one recent approach). This is especially important given that the spatial distribution of targets varies not only between conditions and studies, but also within a single episode of foraging, as each consecutive target is found and removed and the distribution becomes sparser.

Our proposed approach

We propose an alternative strategy for analysing visual foraging data, which involves modelling the process as a generative sampling without replacement procedure, implemented in a Bayesian multilevel model. A strength of our approach is that it leads to measurements (model parameters) that are independent of the number of targets presented on screen and the number of targets that the participants are tasked with finding. In addition, we can estimate the latent biases that led to the run statistics.

We begin by demonstrating the modelling approach for a single trial of visual foraging data from [11], a classic study from the animal foraging literature. From this single trial, we can estimate a parameter that represents the probability that a forager will stick to the same target type or switch to a different type. We will then extend the model to multiple trials and conditions, adding in a bias for target type, and show that our model is able to detect small differences in these parameters that would not be picked up using the dependent variables of maximum run length or number of runs. We then extend to multi-level models to incorporate individual differences, and demonstrate the use of these models using previous data from [32].

Finally, we present a version of the model that incorporates a proximity bias, allowing us to estimate how likely participants are to select a target close to the previously selected target. We use this to demonstrate how the proximity bias affects the calculation of the other biases using example data from [21].

Pulling balls out of a bag

We are modelling visual foraging as a process of weighted sampling without replacement, in what we call a ‘bag foraging model’. Targets are selected one at a time and we define t(i) = t as the ith target selected in a trial, i = 1, 2, …. Our model is binary: targets can be one of two classes, t ∈ {a, b}, with n and n counting the number of each class of targets that have not yet been selected. n = n + n is the total number of eligible targets remaining in trial. We assume that any biases in target selection strategy are constant within, and over, trials.

Single trial

Materials and methods

We start with only one trial of data and assume targets of class a and b are equally likely be selected. We model the preference, p ∈ [0, 1], for selecting a target of the same class as the previously selected target. Higher values of p will lead to longer runs with the same class of target being selected repeatedly, while values of p < 0.5 indicate a preference for alternating between target classes.

For the first target selection, t1, the probability of selecting a target of class a, is simply the proportion of a’s present in the stimulus:

For all subsequent target selections, t2, t3, …, the probabilities Pr(t = a) and Pr(t = b) = 1 − Pr(t = a) depend on the the identity of the previously selected target, t:

An easy way to think about p is to consider the special case when there are equal numbers of both classes of targets left. In this case, p can be interpreted as:

The model was coded in Stan and we use logit link functions to ensure that our probabilities will lie between 0 and 1. I.e., we fit b = logit(p). We define our prior as b ∼ N(0, 1), which translates to a prior prediction that p will fall within 0.1 and 0.9 approximately 95% of the time. The code is presented in S1 Appendix and is available on the Open Science Framework (https://osf.io/7yuaz/). We ran the model using R (v4.0.3) and rStan, v.2.21.2, [33] with four chains and 1000 iterations. Further details of model checking procedures can be found in S1 File).

Chicken foraging

We imported chicken foraging data from [11], which is often referenced as one of the classic papers in the foraging literature. The data is from one chick, choosing between two types of food (grains of rice), and is shown in Fig 1.

After running the model on the data, we inspected traceplots and model summary statistics (, neff) to check that for successful convergence. These can be found in S1 File. The posterior is shown in Fig 3. By calculating the 53% and 97% HPDIs (highest posterior density intervals) around the preference to select the same target class, we obtain intervals of [0.80, 0.85] (more likely than not) and [0.73, 0.88] (very likely). We use these intervals over the more common, but equally arbitrary, 95% for the following reasons: i) 53% represents a little over half of the distributions while 97% represents most of it. ii) there are many cognitive biases associated with the interpretation of probabilities and it is possible that 50% intervals could be viewed as being more likely to contain the “true” value than not. iii) 97% offers a more conservative choice than 95% while avoiding numerical difficulties in obtaining stable estimates out in the tails of the distribution.

Fig 3

Prior and posterior probability distributions for p for the chicken foraging data from Dawkins (1971).

Stable estimates over trial length and number of targets

How much data is required to get a stable estimate of these biases? [11] only looked at the first 100 of 2000 targets found (5%). We ran a simulation where the chickens found greater proportions of the grains (10% or 100%) and found that our model does improve with more data, but only marginally: there is no great benefit to a 20-fold increase in grains in this simple simulation. We also ran a simulation where the chickens always found 100 grains, but the total number of grains varied, and again found this had only small effects on the estimates of p. Our method appears to offer robust model fits even with relatively small numbers of targets (see S1 File).

Multiple trials & conditions

One of the issues with previous measures used in the foraging literature is that they are not able to address why run lengths may differ. That is, what are the underlying cognitive processes that drive a switch to a different target? Our method allows us to consider multiple biases: for example, in addition to a forager’s tendency to stick with one target type or to switch (p) we can add in a second parameter, p, which reflects the bias towards one target type or another (e.g. if one target has a higher value). Increasing either or both parameters will lead to a smaller number of longer runs, but for different reasons.

We therefore expanded our model to work over multiple trials and conditions, and allowed participants to have a preference for one target class over the other:

We estimate p and p independently for each condition, under the assumption that they are constant over trials from the same condition. Our model then becomes:

See S2 Appendix for the Stan code; the model was run in a similar manner to the one trial model.

Misattribution example: A p bias with no p bias

A strength of our model is it allows us to more accurately determine the biases that underlie foraging patterns. We can compare a condition with no biases (p = 0.5 and p = 0.5, a ‘neutral’ condition) with one where the observer prefers one target over the other (p = 0.8 and p = 0.5, which we call a ‘target bias’ condition, as in this case one target is preferred over the other e.g. because it has been given a higher value in the experiment, or because of its visual properties). An ANOVA (the standard statistical method used in foraging papers) can detect a statistically significant difference between these two conditions in terms of both the number of runs (F = 91.1, p < 0.001) and the maximum run length (F = 99.1, p < 0.001): see Fig 4A). However, it is not clear from this analysis what is driving these effects. The posterior probabilities from our model are presented in Fig 4C and clearly distinguish the underlying processes i.e. that there is no difference between p in the two conditions but a clear difference between p: if we repeat the procedure 100 times, we find a difference in p 100% of the time, and a difference in p only 2% of the time.

Fig 4

Results for simulated biases in the bag model.

A: boxplots showing the maximum run length and number of runs in each of our two simulated misattribution conditions. B: boxplots showing the maximum run length and number of runs in each of our two simulated differentiation conditions. C: density plots showing the p and p values calculated by our model for each of the two simulated misattribution conditions. D: density plots showing the p and p values calculated by our model for each of the two simulated differentiation conditions.

Differentiation example: Two conditions which give the same run statistics

To demonstrate the power of our approach, we can look at a simulation example, comparing a condition with a small stick bias (p = 0.5, p = 0.55) to a condition with a small target type preference (p = 0.65, p = 0.5). We simulated 50 trials from an experiment with these two conditions, with each trial consisting of 40 targets. As can be seen in Fig 4B, these two conditions give rise to very similar distributions of maximum run lengths and number of runs. Indeed, when applying the standard ANOVA to these data, we fail to detect a statistically significant difference: for example, in one simulated dataset, F = 0.02, p = 0.9 for the maximum run lengths and F = 0.41, p = 0.52 for the number of runs.

In contrast, running our model on these data results in the posterior probability distribution seen in Fig 4D. From these, we can calculate the difference between conditions, which can be summarised with 97% HPDIs for p, [−0.09, −0.02], and p, [0.10, 0.17]. These intervals do not cross 0, consistent with a difference between conditions for each measure separately. An alternative way to summarise these posterior distributions is to calculate Pr(difference > 0|data). We can then carry out a power analysis by repeating this procedure 100 times. When we do this, we detect a significant difference in p and p in 100% and 83% of simulations, respectively. In contrast, applying the same power analysis using the standard run statistics, we find a statistically significant difference in the number of runs in just 8% of the simulations. Similarly, we find a difference 21% of the time for the maximum run length statistic. We detect a difference in either statistic, and hence detect a difference between conditions, only 29% of the time. Our approach is therefore not only more sensitive than run statistics to the difference between conditions, but is able to correctly attribute the difference to a combination of a bias to switch and a preference for one target over the other.

Multi-level: Tagu and Kristjánsson (2021)

The final version of the bag foraging model allows multiple participants in a multi-level framework (see S3 Appendix). Both p and p are allowed to vary from one participant to the next.

Using simulated data, we can show that our model can recover both overall p and p biases, as well as the parameters from individual participants (see S1 File for details).

Our model also works well with more complex examples. In a recent study, [32] included a target value manipulation, where participants searched for three target types (among three different types of distractors). Participants had to earn a certain number of points in order to complete the trial, and in the ‘value’ condition, one of the target types was deemed high value, whereas the other two were low value. We would therefore expect this condition to have a higher p bias than the ‘no value’ control, where all targets had the same value. This is indeed what we see in Fig 5. This example also shows the flexibility of the model; in this experiment, the proportion of the different target types (high and low value) was not equal to 50:50, and trials were not necessarily equal lengths. Note, for simplicity, we considered only the blocks where participants used a computer mouse in our analysis, and only the trials where the targets had a random distribution (see S1 File).

Fig 5

p and p biases across participants in multi-level re-analysis of Tagu and Kristjánsson (2021), showing both the value and no-value conditions.

Discussion

The bag foraging model we have presented above can account for run behaviour in foraging tasks. Alongside data from [11] and [32], we have also fit the multi-level model to data from [17] (presented in S1 File) and can replicate their key findings. For example, on average, participants have a lower p in the feature condition than in the conjunction condition, demonstrating that they are more likely to have longer ‘runs’ of one target type in the conjunction condition. We can also clearly identify the so-called ‘superforager’ participants, who show a much smaller gap between conditions. However, our method also allows us to see new patterns in the data. For example, the p values in the conjunction condition were often below 0.5. This indicates that participants had a preference for the targets labelled 0, suggesting that by chance, in this dataset, these targets were more salient: S1 File shows how the individual pz correlated well with the participants’ tendency to select a ‘0’ for the first target on each trial. Our method therefore allows deeper analysis of the underlying biases than has been possible with the metrics used in previous studies, while also correlating well with those measures.

One limitation of our model is that it can give less accurate parameter estimates if there are very extreme biases e.g. if the two target types are found almost exclusively in disjoint, non-overlapping runs. If p = 1 (and p = 0.5), our simulated results suggest that the recovery of the p parameter will be accurate, but the posterior for p is shifted away from 0.5. If we take the converse example where p = 1 (and p = 0.5), we obtain a model fit that generates posterior predictions that are identical to the types of runs that would be generated by these parameters i.e. the model correctly predicts that all the targets of one type will be selected first. However, we see that both p and p estimates are actually close to one (see S1 File). This is because an extremely strong target preference means that the participant will find all these preferred targets before the non-preferred, generating a trial where they appear never to switch. Thus, the estimated parameters are entirely consistent with the observed data. However, assuming a slightly less extreme value of p = 0.97, or adjusting the priors, allows better recovery of the correct parameters (see S1 File). In the case of the [17] data, the p bias appears to be real, as manual inspection of the first five target selections per trial in the conjunction condition shows a similar bias towards the target labelled ‘0’. Thus, some caution should therefore be exercised in interpreting parameters where there are strong biases, but it is simple to check if there are model fitting problems.

Spatial biases

Materials and methods

Our initial model incorporates several biases which appear to help explain the findings in previously published research [32]. However, this model has no representation of the spatial location of targets. It seems plausible that people are more likely to select targets nearby to the one they have just selected. Proximity could be particularly important in feature search, where the targets are relatively easily distinguished so there are fewer additional constraints on selection; if so, perhaps proximity is relatively less important for conjunction search. We can also investigate the effect of direction/momentum: do observers incorporate some form of momentum to their target selections, or do they prefer to double back on themselves? Extending our bag foraging model to incoporate spatial biases allows us to see to what extent these hypotheses are true, and how the inclusion of a spatial bias will affect our estimation of the other biases.

For our second model, we take our stimuli to be a collection of targets, t(i) = (t, x, y), where (x, y) gives the target’s location and t ∈ {a, b} gives the target’s class. For mathematical convenience, we will take a and b to factor levels with a = 1 and b = −1. If we take d(i, j) to be the Euclidean distance between targets (x, y) and (x, y), we can define a proximity measure between the two items as: where σ is a proximity tuning parameter that controls how quickly ρ decreases from 1 to 0.

Similarly, we can define the angular difference θ(i, j) to be: where atan2(j, i) is the direction of travel from i to j and f(ϕ1, ϕ2) calculates the angular difference:

This should give a value between 0 and 1, with 0 indicating a target that “in front” of the current target, 0.5 represents a change of direction of 90° and 1 represented doubling back in the direction we just came from. Having calculated θ, we can now define:

In the bag foraging model, all targets of the same class were interchangeable and so a simple Bernoulli process could be used to model each target selection. However, this no longer holds, as each target has a unique spatial location that needs to be taken into account. Therefore we will use a categorical model with the weights for each remaining target being defined as: where g is the inverse logit function and m is an indicator function that equals 1 if t = t and -1 otherwise. Targets that have already been selected earlier in a trial have their weights set to 0. S4 and S5 Appendices show the code for the basic and multi-level spatial models, respectively.

Results

Simulation examples

We present a simple example using simulated data in S2 File, and demonstrate that we can recover the parameters used to generate the data. A more interesting example can be seen in the case where the targets have spatial structure in the environment, being found in patches (See S2 File for example stimulus). In our simulation, we set b and b to be zero, and used a proximity tuning value of 15. If analysed without including a proximity bias parameter, the model mistakenly generates a b bias: the clumping of targets does indeed lead to them being collected in runs, simply because our simulated participant has a preference for nearby targets (see Fig 6). However, the model including the proximity bias parameter is able to recover the correct parameters. Our model is therefore able to distinguish between a preference for a similar target and a preference for a nearby target, a distinction that is likely to be particularly important in more naturalistic set ups where target clumping might be expected to be prevalent.

Fig 6

Posterior distributions for both a bag foraging model and a spatial model trained on patchy stimuli, where target types are clumped.

The shaded ribbons for the proximity and direction weighting indicate 53%, 89% and 97% HPDIs. The spatial model gives more weight to nearby targets (i.e. where the distance is close to zero), and has no particular preference for any direction, as indicated by the fact that a value of 1 falls within the HPDI for all directions.

Multi-level: Clarke et al. (2018)

The data from [21] closely replicates [17], yet has around four times as many participants, giving us greater power to investigate individual differences in feature and conjunction foraging. The fixed-effects of the model are shown in Fig 7. There is a small bias towards repeating target selections during feature search, and this increases to approach ceiling in the conjunction condition. There is also a very strong proximity bias in both conditions, with very little weight given to targets more than a quarter stimulus width from the currently selected target. The parameter for the direction bias comes out as negative, leading the model to favour doubling back on itself to select targets that may have been skipped over (see S2 File for simulated examples of how different direction biases affect foraging behaviour). Interestingly, both of these biases are slightly weaker for conjunction foraging, suggesting participants are relying less on spatial attention and emphasising feature-based attention.

Fig 7

Posterior distributions for our model when trained on data from Clarke et al. (2018).

The shaded ribbons for the proximity and direction weighting indicate 53%, 89% and 97% HPDIs. Note that there is a higher weighting for directions closer to pi i.e. completely reversing the direction, indicating negative momentum.

As this model was fit using a multi-level framework, we can also look differences between participants. Fig 8 shows how the parameters vary over participants and conditions. We can see that with the exception of parameters linked to the spatial biases, there are little-to-no correlations between different parameters across conditions, or within parameters across conditions. However, the two spatial parameters appear to be correlated: participants with a stronger preference to select nearby targets also have a more negative direction bias. Participants with a weaker proximity bias are more likely to have a positive directional bias. While there is no clear partition of participants into subgroups, the two opposing ends of this range could be interpreted as a strategy to search within local patches compared to a more global scanning strategy.

Fig 8

Individual differences in Clarke et al. (2018).

The first two rows show the correlations in the random effect structure between different parameters. We can see that p and p appear to be independent from one another, and independent from the spatial biases b and p. However, b and p are correlated with one another in both the feature and conjunction conditions. The bottom row shows the correlation between conditions for each parameter in our model.

Finally, we can use the fitted model to simulate new data and compare run statistics and inter-target distances with the original empirical data. We do this for n = 100 samples from the fitted posterior probability distributions so that we can quantify the uncertainty in the model’s predictions. The results (Fig 9) show a good correspondence between the posterior predictions and empirical data. Both the summary run statistics (the two statistics most commonly analysed in the visual foraging literature) are a close-to-perfect match between the model and the empirical data. The inter-target distances are more interesting. The model offers a reasonably good fit, especially in terms of the mean distances and in the later half of a trial. However, we can see two main differences: i) the variance in the human data is larger, suggesting that perhaps the negative exponential proximity function down-weights more distant targets too much; and ii) there is an interesting dynamic at the start of the trial in which the first few targets selected tend to have shorter inter-target distances than the model would expect. This raises the possibility that participants either chose an initial target that is in a relatively dense patch of potential targets, or that they are able to accurately plan out an efficient order for the first five or so target selections that minimises the distance that they would be required to travel. This observation highlights another use of the model, which is to detect deviations in human behaviour from the model’s predictions. These deviations can guide the direction of further investigation.

Fig 9

Posterior predictions for Clarke et al. (2018).

(top:) Scatter plots between empirical and predicted summary run statistics. Error bars indicated 89% HPDI for the posterior predictions. (bottom:) Comparison of inter-target distances and how they vary over time. Shaded region gives 89% intervals.

The spatial model presented extends the simple bag foraging model to incorporate new parameters that further explain participant behaviour on foraging tasks by accounting for spatial attention. This section demonstrates the flexibility and modularity of the bag foraging model, which can be used as a starting point for exploring many questions about foraging behaviour. As well as presenting the model fit for the [21] dataset above, S2 File presents fits for a number of other datasets, including [17], [32] and [28]. Another key strength of our model demonstrated in S2 File is that it is able to generate good parameter estimates with relatively little data (e.g. one trial per participant per condition), in stark contrast to using traditional run statistics, which reduces each trial to one summary number. This means that our model could allow analysis of foraging data from populations where it is difficult to carry out large numbers of trials, such as from clinical populations or when studying animal behaviour.

General discussion

One of the biggest strengths of our approach is its relative simplicity. By taking the task structure of sampling without replacement into account, we need only four parameters to characterise the relatively complex sequences of runs and inter-target distances produced during foraging tasks. While approximating these parameters with summary statistics (e.g. the mean of the run lengths) in many cases will lead to broadly similar conclusions, we have demonstrated that in some instances, run statistics give misleading results that our approach avoids. In addition, summary statistics require much more data per participant and condition to be able to make good estimates. By treating each selection instance as a separate observation, our approach reaches stable estimates of biases far more efficiently.

Our model is also extremely flexible, and could be adapted in many ways. The underlying assumptions of the model are all tuneable and testable: for example, it would be straightforward to compare and contrast different proximity functions, as our choice of using an isotropic negative exponential is relatively arbitrary, and other functions may offer improved performance for predicting behaviour. The momentum function also assumes a monotonic trend from going forwards in a straight line to going backwards, but again, there are other possibilities that could easily be tested. For example, it could also be explored whether functions that allow for a forward-backward preference over perpendicular movement are better. Interesting further questions could explore momentum and proximity biases across different modalities. Most of the data included here was based on finger or mouse-click foraging, but larger movements (such as moving around a large area) or smaller ones (such as eye movements) are likely to be characterised by different proximity and momentum functions.

More broadly, the model is extremely straightforward to extend. As it stands, our model works well with different types of foraging tasks, as we have demonstrated using both exhaustive and non-exhaustive foraging example datasets. However, while outside the scope of the current paper, it would be easy to also implement stopping rules for models of non-exhaustive foraging, and it would be possible to make quantitative predictions about distributions by assuming different stopping rules (for example, by assuming that after each target selection, there is some given probability to terminate the trial). This would allow modelling of patch-leaving rules, assuming that one trial acts as one patch. Within a trial, it would also be possible to model patches made up of multiple “clumped” targets on a single large display, by modifying the spatial bias component to include a categorical factor to indicate the different clumps of targets (that are therefore assumed to be perceptually ‘further away’ from each other). The p bias could also be broken down into sub-biases based on how preference changes based on stimulus properties such as colour, luminance and value. The model could also be extended to explore how biases change temporally (e.g. over multiple trials), or to include new biases, such as a bias for the initial target selection. Finally, it would also be possible to extend beyond two target categories: for example, for three categories, instead of fitting one value for Pr(t = a) and and then letting Pr(t = b) = 1 − Pr(t = a), we would need to add an extra parameter, which could be parameterised Pr(t = c) = 1 − Pr(t = a) − Pr(t = b).

In other aspects of foraging, such as the marginal value theorem for predicting when foragers should leave a patch [9], formal mathematical modelling has been used to effectively represent the underlying structure of foraging behaviour. Our work provides a similar framework for target selection during foraging, extending our theoretical models of foraging behaviour and giving a way for researchers to begin to understand the cognitive processes that underlie performance in these tasks. We envision future extensions bringing these and other models together into a complete understanding of all aspects of foraging. While a full account of human foraging is of course likely to be extremely complex, involving many factors [10], we are optimistic that our approach will provide a framework to begin this challenging task. By making our code freely available, we hope to enable other researchers to update the model as they require, perhaps to take into account other cognitive factors that they think may be important in foraging tasks, or to enable them to use the model to study their own variants of the task.

There are a number of potentially important factors that we do not consider in the current model. For example, it does not currently allow for modelling inter-target selection times, which are thought to be important in determining foraging behaviour [28]. Similarly, it does not explicitly model the effect of background clutter or distractors, although implicit effects (such as decreasing the salience of one of the target classes) will be captured. As most foraging studies to date have not focused on distractor effects, information on the distribution of distractors is not routinely included in freely-available datasets, making it more difficult to model this aspect of the task. However, if such data were available, the model could be extended to treat distractors in the same way as targets, but with a sampling-with-replacement and inhibition of return process for when a distractor is selected.

Conclusion

Our Bayesian statistical model of visual foraging provides a simple and flexible way to estimate cognitively meaningful parameters underlying foraging tasks. We demonstrate that the model is able to reproduce the patterns found in existing data, and also provides new avenues for research questions, such as the role of salience in predicting target preferences and how proximity and direction biases may represent stable individual differences across different task variants. By considering the task structure in developing our model, we can go beyond simple linear models to generate more powerful predictions and deeper insights into foraging behaviour.

Supplementary materials part 1.

Supplementary materials for the bag foraging model.

(PDF)

Click here for additional data file.

Supplementary materials part 2.

Supplementary materials for the spatial foraging model.

(PDF)

Click here for additional data file.

Bag foraging model—One trial.

The Stan code for the basic version of our sampling-without-replacement model for visual foraging (may be opened in R or in a text editor).

(STAN)

Click here for additional data file.

Bag foraging model—Multiple trials and conditions.

The Stan code for the version of our sampling-without-replacement model for visual foraging, incorporating multiple trials and conditions (may be opened in R or in a text editor).

(STAN)

Click here for additional data file.

Bag foraging model—Multi level.

The Stan code for the version of our sampling-without-replacement model for visual foraging, incorporating a multi-level random effects structure (may be opened in R or in a text editor).

(STAN)

Click here for additional data file.

Spatial foraging model.

The Stan code for the version of our sampling-without-replacement model for visual foraging, incorporating spatial biases (may be opened in R or in a text editor).

(STAN)

Click here for additional data file.

Spatial foraging model—Multi level.

The Stan code for the version of our sampling-without-replacement model for visual foraging, incorporating spatial biases and a multi-level random effects structure (may be opened in R or in a text editor).

(STAN)

Click here for additional data file.

6 Dec 2021

Dear Dr Hughes,

Thank you very much for submitting your manuscript "Building Bayesian Cognitive Models of Visual Foraging" for consideration at PLOS Computational Biology. As with all papers reviewed by the journal, your manuscript was reviewed by members of the editorial board and by several independent reviewers. The reviewers appreciated the attention to an important topic. Based on the reviews, we are likely to accept this manuscript for publication, providing that you modify the manuscript according to the review recommendations.

As you will see from the comments below the reviewers were all very positive and appreciative of the effort put into thie work. We expect this to be a minor revision.

Please prepare and submit your revised manuscript within 30 days. If you anticipate any delay, please let us know the expected resubmission date by replying to this email.

When you are ready to resubmit, please upload the following:

[1] A letter containing a detailed list of your responses to all review comments, and a description of the changes you have made in the manuscript. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out

[2] Two versions of the revised manuscript: one with either highlights or tracked changes denoting where the text has been changed; the other a clean version (uploaded as the manuscript file).

Important additional instructions are given below your reviewer comments.

Thank you again for your submission to our journal. We hope that our editorial process has been constructive so far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Ulrik R. Beierholm

Associate Editor

PLOS Computational Biology

Natalia Komarova

Deputy Editor

PLOS Computational Biology

***********************

A link appears below if there are any accompanying review attachments. If you believe any reviews to be missing, please contact ploscompbiol@plos.org immediately:

[LINK]

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: The paper Building Bayesian Cognitive Models of Visual Foraging by Clarke et al. introduces a new approach to modeling data from visual foraging tasks. Importantly, the approach can disentangle different influences on target selections and predict e.g. run-behavior in foraging for different conditions. The model is a much-needed tool for a field that lacks adequate methods to deal with the rich data produced by foraging tasks.

The paper is well structured, the rationale is clear, and the model and its applications are presented in a comprehensible manner. I only have a few minor points and suggestions, which I list below (not necessarily in order of importance).

* p2: "... below the average rate of return across all patches". - I suggest adding something to highlight that the avg. ROR across all patches also includes travel time during which no elements can be collected. Sometimes people not familiar with MVT are confused about this ...

* p2: The idea of a "search image" is used on p2. Perhaps after this passage, you could mention the related concepts of attentional templates concerning visual search in humans. The concept of templates appears later in the current manuscript but was nowhere introduced ... this might be a possibility.

* On p2 you suggest that foragers can keep "multiple target templates active in working memory". On p3 you write " potentially implying that they are able to hold multiple target templates simultaneously in working memory", as a special case for super-foragers, implying that normally this might not be possible (somewhat in conflict with what was said on p2). I think the topic is still debated. The point here could probably be made without committing to either view.

~ p4: It does not become clear that (and why) the model focuses on a binary case with two target types. Also, in the discussion, where limitations are discussed, this point is not taken up either.

* On several occasions, you point out that existing methods might not adequately model the underlying cognitive processes, implying that your approach does. At one location, it is also explicit: "A strength of our model is it allows us to more accurately determine the cognitive processes underlying foraging patterns." While I completely agree that your model is a significant step forward compared to earlier methods and goes a long way in disentangling different biases, I would say that it still does not "determine the cognitive processes underlying foraging patterns." How do these processes lead to a particular switching frequency? Why is there a preference for certain features? ....

* p8: Suddenly, the term "bag model" appears. It also occurs as "bag foraging model", sometimes set in italics. Perhaps this term should be introduced earlier.

* p5: Figure 3 refers to "salience" (and it comes up later again). The main text does not introduce the idea that salience could drive a target tepy preference. The real-data example uses the value as the reason for preferences ...

* p5: Figure 2: "the chicken data" sounds a bit odd. Perhaps add a reference?

* p7: I find the term "conflated power example" unclear.

* p9: "identical to the training data": I think the ML framing (which only appears here) could be confusing for some readers

* p11: "leading the model to favor doubling back on itself to select targets.". I have only a vague idea of what this means ... perhaps a figure could be helpful here.

* p11: what is an "eligible target". Aren't all targets eligible?

* p15 (an also the title): "Bayesian cognitive model" - I know this term is used for different things (and e.g. in the Lee and Wagenmakers book in a similar context as here). Personally, I try to avoid it because people easily confuse it with Bayesian models of cognition, i.e. the idea that cognitive processes follow Bayesian principles. If I see it correctly, your study does not suggest this (of course, it might also not exclude it) but simply uses a Bayesian statistical framework.

* The paper contains many typos. Here are some I spotted:

- p1, 1st para: missing closing parenthesis

- Figure 1: Caption says "top" and "bottom," but panels are on the left and right.

- Figure 1: I think "first three trials" should be "first three selections"

- p5: HDPIs --> HPDIs, also the acronym is never introduced; in Figure 5 you use "HDI".

- p5: Footnote: consider removing point iv). People not very familiar with Bayesian stats might think this might really be of relevance

- p6: "(5%" - closing parenthesis missing

- p6: "our model__s__ then becomes"

- p7: "seen in 3" ... "Figure" is missing

- "super-foragers" vs. "superforagers"

- P10: "Figure Figure 5"

- p11: "to appear to be"

Reviewer #2: I was impressed with this manuscript. It is written very clearly and efficiently and presents to me what seems like a very useful approach to understanding foraging. While I do think that accounts based on run behavior can still definitely aid our understanding of foraging in particular and visual attention and attentional selection more generally, the current approach has considerable merits, which the authors make a very clear and concise case for.

I therefore do not have many comments, and my comments therefore can be taken as food for thought.

Firstly, I would recommend not talking about "confounds" with regard to the literature that has focused mostly on selection runs. Yes, it is true that run length, run number, switch rates are interdependent and correlated, but I do not see the problem that the word confound implies. As I see this, the current authors sampling approach can well hold it’s own without trying to imply that run based approaches need to be avoided as language such as confounds certainly implies. The uncertainty with regard to cognitive mechanisms behind run behavior is something that the authors could focus more in in my opinion – for example the finding that children under 12 tend not to switch, may point towards mechanisms.

When the authors talk about how optimal foraging theory has sometimes run into problems, I thought they should mention a recent finding by Kristjánsson, Björnsson & Kristjánsson (https://doi.org/10.3758/s13414-019-01941-y ) where the authors claim that human foraging in paradigms in many ways identical as those under investigation in the manuscript here deviated from the predictions of optimal foraging conceptions and that participants kept on foraging within the same patch much longer than expected. The authors concluded that human foraging is probably influenced by too many factors to be captured with a relatively simple mathematical model. This actually speaks to what my main point about this paper is in that there are so many factors at play that determine foraging performance that capturing them all within a single model is probably premature, if not impossible.

With regard to this issue of „patch leaving“, I encoourage the authors to provide a little more detail and be a bit more explicit in the discussion about how this could be implemented within their model. This is only briefly metnioned in the current discussion.

I went through the mathematical modelling and it seemed to be very solid.

I do share the authors enthusiasm with regard to cluttered scenes. It would for example be highly interesting to vary noise levels – and in fact many means are possible it seems.

Overall, I commend the authors on what is in my opinion a very fine contribution to the literature.

I was impressed with this manuscript. It is written very clearly and efficiently and presents to me what seems like a very useful approach to understanding foraging. While I do think that accounts based on run behavior can still definitely aid our understanding of foraging in particular and visual attention and attentional selection more generally, the current approach has considerable merits, which the authors make a very clear and concise case for.

I therefore do not have many comments, and my comments therefore can be taken as food for thought.

Firstly, I would recommend not talking about "confounds" with regard to the literature that has focused mostly on selection runs. Yes, it is true that run length, run number, switch rates are interdependent and correlated, but I do not see the problem that the word confound implies. As I see this, the current authors sampling approach can well hold it’s own without trying to imply that run based approaches need to be avoided as language such as confounds certainly implies. The uncertainty with regard to cognitive mechanisms behind run behavior is something that the authors could focus more in in my opinion – for example the finding that children under 12 tend not to switch, may point towards mechanisms.

When the authors talk about how optimal foraging theory has sometimes run into problems, I thought they should mention a recent finding by Kristjánsson, Björnsson & Kristjánsson (https://doi.org/10.3758/s13414-019-01941-y ) where the authors claim that human foraging in paradigms in many ways identical as those under investigation in the manuscript here deviated from the predictions of optimal foraging conceptions and that participants kept on foraging within the same patch much longer than expected. The authors concluded that human foraging is probably influenced by too many factors to be captured with a relatively simple mathematical model. This actually speaks to what my main point about this paper is in that there are so many factors at play that determine foraging performance that capturing them all within a single model is probably premature, if not impossible.

With regard to this issue of „patch leaving“, I encoourage the authors to provide a little more detail and be a bit more explicit in the discussion about how this could be implemented within their model. This is only briefly metnioned in the current discussion.

I went through the mathematical modelling and it seemed to be very solid.

I do share the authors enthusiasm with regard to cluttered scenes. It would for example be highly interesting to vary noise levels – and in fact many means are possible it seems.

Overall, I commend the authors on what is in my opinion a very fine contribution to the literature.

Reviewer #4: This paper describes two versions of a model of visual foraging aiming at providing more details about foraging behaviour than the run statistics often used in the literature. Notably, run statistics can inform about the tendency to stick with one target type or to switch, while the “pulling balls out of a bag” model offers the possibility to assess pA (the preference for one target type or another) on the top of pS (the tendency to stick with one target type or to switch). The second version of the model allows assessment of spatial biases with two additional parameters for proximity and direction.

I enjoyed reading this paper. It is nicely written, and the paper is well motivated and of high interest in the field of visual foraging. The maths is explained in a very pedagogical way so that it is also understandable by non-experts. Moreover, I wish to thank the authors for providing useful, detailed and well-constructed supplementary materials on the top of the main article.

I only have a few comments below that I hope are useful.

Jerome Tagu

1. I find the four parameters described in the paper useful for assessing foraging behaviour, but I feel they are presented independently, in separate analyses (especially the spatial biases that seem separate from the selection biases). I wonder whether it would be possible to have the four parameters in a single analysis, or equation (a bit as we would have in a regression analysis), so that the behaviour of a given individual could be summarised in a very straightforward way (e.g. this participant’s behaviour is explained at 80% by a preference for target A and at 20% by a preference for proximity). NB: I’m not an expert in Bayesian modelling, so maybe this is something not doable, but if doable, it would be very cool.

2. It is said multiple times (l.84-87, l.306-311, and probably elsewhere), that the model can generate good parameter estimates with little trials. Is the model also resistant to the manipulation of the number of target selections per trial? Knowing this would be very useful for the study of populations such as children, who have difficulties to perform many target selections per trial (see, e.g. Olafsdottir et al., 2016, 2020).

3. I understand that the model is somehow limited by the number of target types (can only be used with two target types). More and more studies are now contrasting 3+ target types, and I would find useful to update the model so that it could be used to assess the relative preference (pA) for n = 3 or more target types.

4. In the re-analysis of the data from Kristjansson et al. (2014), the model allows a clear identification of “super-foragers” who show low pS in conjunction foraging (e.g. Fig.21 of the supplementary materials part 1). Just to be sure: are these super foragers the same individuals as the ones identified as super foragers in the original analysis? Similarly, are the participants showing a strong preference for valuable targets on supplementary figure 26 the same individuals as the ones identified in the original analysis in Tagu & Kristjansson (2022)?

5. Data used for testing the model were imported from studies involving “exhaustive” foraging (e.g. Kristjansson et al., 2014), where participants were required to select all available targets, as well as from studies involving “non-exhaustive” foraging (Dawkins, 1971; Tagu & Kristjansson, 2022), where participants only select some of the available targets. If I understand correctly, the model doesn't seem to be influenced by the differences in task instructions and can reliably generate pS and pA in both conditions. Maybe the authors would like to highlight this other strength of the model, on the top of its other qualities.

Specific comments:

- Figures: the figures often have multiple panels and which panel is presented in the text is not always very clear. Please consider labelling the panels with letters.

- l.8-9: a parenthesis seems to be missing here

- l.223-227: could the authors indicate which target type is target 0? The same colours (red, green, yellow, blue) are often used by different foraging studies. Is the preference for target type 0 a reliable preference for a given colour (e.g. red) found in several datasets, or is it specific to the dataset that is analysed here?

- l.261: typo, “figure figure 5”

- l.271-272: I must be missing something here, but the text is referring to a negative direction bias while in figure 6 I (mistakenly?) see a positive direction bias.

- Supp. Materials part 1, Figure 9: as the curves are similar, it is difficult to identify the four distributions. Maybe using line colour instead of filling colour in the legend would help identify the curves on the plot. Same thing for Fig.3 of supplementary materials part 2.

- Supp. Materials part 1: a small bias on pS is tested with a preference of .6 (end of p. 8), while a small bias on pA is tested with a preference of .8 (p. 9). But is .8 really a “small” preference? Why not using a bias of .6 in both analyses?

- Supp. Materials part 1, end of p. 17: it is said that the experiment contains 20 trials per condition, and then that the analysis was restricted to trials 5-25. So there appears to be a typo somewhere here!

**********

Have the authors made all data and (if applicable) computational code underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data and code underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data and code should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data or code —e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: None

Reviewer #2: Yes

Reviewer #4: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #4: Yes: Jérôme Tagu

Figure Files:

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

Data Requirements:

Please note that, as a condition of publication, PLOS' data policy requires that you make available all data used to draw the conclusions outlined in your manuscript. Data must be deposited in an appropriate repository, included within the body of the manuscript, or uploaded as supporting information. This includes all numerical values that were used to generate graphs, histograms etc.. For an example in PLOS Biology see here: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5.

Reproducibility:

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

References:

Review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript.

If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

22 Dec 2021

Submitted filename: foraging_responsetoreviewers.docx

Click here for additional data file.

6 Jan 2022

Dear Dr Hughes,

We are pleased to inform you that your manuscript 'Foraging as sampling without replacement: a Bayesian statistical model for estimating biases in target selection' has been provisionally accepted for publication in PLOS Computational Biology.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch with a set of requests.

Please note that your manuscript will not be scheduled for publication until you have made the required changes, so a swift response is appreciated.

IMPORTANT: The editorial review process is now complete. PLOS will only permit corrections to spelling, formatting or significant scientific errors from this point onwards. Requests for major changes, or any which affect the scientific understanding of your work, will cause delays to the publication date of your manuscript.

Should you, your institution's press office or the journal office choose to press release your paper, you will automatically be opted out of early publication. We ask that you notify us now if you or your institution is planning to press release the article. All press must be co-ordinated with PLOS.

Thank you again for supporting Open Access publishing; we are looking forward to publishing your work in PLOS Computational Biology.

Best regards,

Ulrik R. Beierholm

Associate Editor

PLOS Computational Biology

Natalia Komarova

Deputy Editor

PLOS Computational Biology

***********************************************************

20 Jan 2022

PCOMPBIOL-D-21-01844R1

Foraging as sampling without replacement: a Bayesian statistical model for estimating biases in target selection

Dear Dr Hughes,

I am pleased to inform you that your manuscript has been formally accepted for publication in PLOS Computational Biology. Your manuscript is now with our production department and you will be notified of the publication date in due course.

The corresponding author will soon be receiving a typeset proof for review, to ensure errors have not been introduced during production. Please review the PDF proof of your manuscript carefully, as this is the last chance to correct any errors. Please note that major changes, or those which affect the scientific understanding of the work, will likely cause delays to the publication date of your manuscript.

Soon after your final files are uploaded, unless you have opted out, the early version of your manuscript will be published online. The date of the early version will be your article's publication date. The final article will be published to the same URL, and all versions of the paper will be accessible to readers.

Thank you again for supporting PLOS Computational Biology and open-access publishing. We are looking forward to publishing your work!

With kind regards,

Livia Horvath

PLOS Computational Biology | Carlyle House, Carlyle Road, Cambridge CB4 3DN | United Kingdom ploscompbiol@plos.org | Phone +44 (0) 1223-442824 | ploscompbiol.org | @PLOSCompBiol