Does intraspecific competition promote variation? A test via synthesis

Abstract

Competitive diversification, that is, when increasing intraspecific competition promotes population niche expansion, is commonly invoked in evolutionary studies and currently plays a central role in how we conceptualize the process of adaptive diversification. Despite the frequency with which this idea is cited, the empirical evidence for the process is somewhat limited, and the findings of these studies have yet to be weighed objectively through synthesis. Here, we sought to fill this gap by reviewing the existing literature and collecting the data necessary to assess the evidence for competition as a diversifying force. Additionally, we sought to test a more recent hypothesis, which suggests that competition can act to both promote and inhibit dietary diversification depending on the degree to which a consumer depletes its resources. The surprising result of this synthesis was that increasing competition did not have a mean positive effect on population-level diet breadth or the degree of individual specialization. Instead, we found that increasing intraspecific competition had a restricting effect on population-level diet breadth in as many cases as it had a diversifying effect. This wide disparity in the effect of competition on consumer diet variation was negatively related to a metric for consumer resource depletion. Altogether, these findings call into question a long-standing assumption of basic evolutionary models and lend some support to recent theoretical predictions. Specifically, these findings support the idea that competition is primarily diversifying for species with a small effect (per unit biomass) on their resources and that resource depletion limits the diversifying effect of competition for consumers with larger ecological effects.

Keywords: adaptive divergence; interaction strength; intraspecific competition; niche width; resource depletion.

Figure 1

Effect sizes suggest no mean effect of competition on population‐level niche width. The log response ratio effect sizes for the studies included in the analysis. Mean effect sizes are coded by the type of study, with dark circles representing experimental studies and lighter triangles representing observational studies. Error bars represent 95% confidence intervals. The summary mean effect size derived from the metafor‐based mixed model analysis is shown at the bottom.

Figure 2

Metrics suggest resource depletion influences population niche width. (A) The relationship between a metric that describes how a consumer population's niche width changes with density, and the degree to which consumers deplete their resources. Values above dashed zero line represent studies where increasing competition among conspecific consumers had a diversifying effect on their population‐level dietary variation. Values below zero represent cases where increasing competition had a restricting effect on dietary variation. Each point represents a single study. Shapes represent different species. The fitted line (red) represents the reduced model including only RD. (B) The more general relationship between niche width and a metric for competition. Both niche width and densities have been normalized to values between zero and one. Shapes and line types represent each species from which the data were derived. The quadratic fitted line (blue) suggests a unimodal relationship and supports predictions of the ICD hypothesis. In both figures, the gray ribbon represents a 95% confidence interval for the simplified fitted lines (red and blue).