Weak interactions, omnivory and emergent food-web properties

Empirical studies have shown that, in real ecosystems, species-interaction strengths are generally skewed in their distribution towards weak interactions. Some theoretical work also suggests that weak interactions, especially in omnivorous links, are important for the local stability of a community at equilibrium. However, the majority of theoretical studies use uniform distributions of interaction strengths to generate artificial communities for study. We investigate the effects of the underlying interaction-strength distribution upon the return time, permanence and feasibility of simple Lotka-Volterra equilibrium communities. We show that a skew towards weak interactions promotes local and global stability only when omnivory is present. It is found that skewed interaction strengths are an emergent property of stable omnivorous communities, and that this skew towards weak interactions creates a dynamic constraint maintaining omnivory. Omnivory is more likely to occur when omnivorous interactions are skewed towards weak interactions. However, a skew towards weak interactions increases the return time to equilibrium, delays the recovery of ecosystems and hence decreases the stability of a community. When no skew is imposed, the set of stable omnivorous communities shows an emergent distribution of skewed interaction strengths. Our results apply to both local and global concepts of stability and are robust to the definition of a feasible community. These results are discussed in the light of empirical data and other theoretical studies, in conjunction with their broader implications for community assembly.

Effect of increasing dietary folate on red-cell folate: Implications for prevention of neural tube defects

Background Recommendations by the UK Department of Health suggest that protection from neural tube defects (NTD) can be achieved through intakes of an extra 400 mu g daily of folate/folic acid as natural food, foods fortified with folic acid, or supplements. The assumption is that all three routes of intervention would have equal effects on folate status.

Methods We assessed the effectiveness of these suggested routes of intervention in optimising folate status. 62 women were recruited from the University staff and students to take part in a 3-month intervention study. Participants were randomly assigned to one of the following five groups: folic acid supplement (400 mu g/day; I); folic-acid-fortified foods (an additional 400 mu g/day; II); dietary folate (an additional 400 mu g/day; III); dietary advice (IV), and control (V). Responses to intervention were assessed as changes in red-cell folate between preintervention and postintervention values.

Findings 41 women completed the intervention study. Red-cell folate concentrations increased significantly over the 3 months in the groups taking folic acid supplements (group I) or food fortified with folic acid (group II) only (p<0.01 for both groups). By contrast, although aggressive intervention with dietary folate (group III) or dietary advice (group IV) significantly increased intake of food folate (p<0.001 and p<0.05, respectively), there was no significant change in folate status.

Interpretation We have shown that compared with supplements and fortified food, consumption of extra folate as natural food folate is relatively ineffective at increasing folate status. We believe that advice to women to consume folate-rich foods as a means to optimise folate status is misleading.

Common Ingroups and Complex Identities: Routes to Reducing Bias in Multiple Category Contexts

We tested the hypothesis that evaluative bias in common ingroup contexts versus crossed categorization contexts can be associated with two distinct underlying processes. We reasoned that in common ingroup contexts, self-categorization, but not perceived complexity, would be positively related to intergroup bias. In contrast, in crossed categorization contexts, perceived complexity, but not self-categorization, would be negatively related to intergroup bias. In two studies, and in line with predictions, we found that while self-categorization and intergroup bias were related in common ingroup contexts, this was not the case in crossed categorization contexts. Moreover, we found that perceived category complexity, and not self-categorization, predicted bias in crossed categorization contexts. We discuss the implications of these findings for models of social categorization and intergroup bias.