Leveraging intelligence for high performance in complex dynamic systems requires balanced goals

Psychometric intelligence is weakly correlated with control performance in micro-worlds. Previous attempts at explaining these low correlations have focused on reliability problems of micro-worlds and/or the need for skills and capabilities not captured in static problem-solving tests. Meta cognitive factors are hypothesized to influence the efficacy of problem-solving capability in managing a micro-world control task, thus explaining some of the low correlations between performance and psychometric intelligence. Specifically, goal level, implicitly the control problem difficulty, is proposed to interact with psychometric intelligence in determining control performance. Forty-six participants managed the micro-world Moro under opaque test conditions. Three-way ANOVA analysis shows an interaction between goal level and intelligence (Raven's APM) for several performance variables over time, indicating the importance of meta-cognitive processes for leveraging psychometric intelligence.