Speed-Accuracy Tradeoffs in Brain and Behavior: Testing the Independence of P300 and N400 Related Processes in Behavioral Responses to Sentence Categorization

Abstract

Although the N400 was originally discovered in a paradigm designed to elicit a P300 (Kutas and Hillyard, 1980), its relationship with the P300 and how both overlapping event-related potentials (ERPs) determine behavioral profiles is still elusive. Here we conducted an ERP (N = 20) and a multiple-response speed-accuracy tradeoff (SAT) experiment (N = 16) on distinct participant samples using an antonym paradigm (The opposite of black is white/nice/yellow with acceptability judgment). We hypothesized that SAT profiles incorporate processes of task-related decision-making (P300) and stimulus-related expectation violation (N400). We replicated previous ERP results (Roehm et al., 2007): in the correct condition (white), the expected target elicits a P300, while both expectation violations engender an N400 [reduced for related (yellow) vs. unrelated targets (nice)]. Using multivariate Bayesian mixed-effects models, we modeled the P300 and N400 responses simultaneously and found that correlation between residuals and subject-level random effects of each response window was minimal, suggesting that the components are largely independent. For the SAT data, we found that antonyms and unrelated targets had a similar slope (rate of increase in accuracy over time) and an asymptote at ceiling, while related targets showed both a lower slope and a lower asymptote, reaching only approximately 80% accuracy. Using a GLMM-based approach (Davidson and Martin, 2013), we modeled these dynamics using response time and condition as predictors. Replacing the predictor for condition with the averaged P300 and N400 amplitudes from the ERP experiment, we achieved identical model performance. We then examined the piecewise contribution of the P300 and N400 amplitudes with partial effects (see Hohenstein and Kliegl, 2015). Unsurprisingly, the P300 amplitude was the strongest contributor to the SAT-curve in the antonym condition and the N400 was the strongest contributor in the unrelated condition. In brief, this is the first demonstration of how overlapping ERP responses in one sample of participants predict behavioral SAT profiles of another sample. The P300 and N400 reflect two independent but interacting processes and the competition between these processes is reflected differently in behavioral parameters of speed and accuracy.

Keywords: N400; P300; SAT; mixed-effects modeling; predictive processing; sentence processing.

Figure 1

Event-related potential (ERP) time-course at Cz (Experiment 1). Shaded regions indicate 83% confidence intervals of the grand mean; non overlap is equivalent to significance at the 5% level. Positivity is plotted upwards.

Figure 2

Topographies of ERPs by condition (Experiment 1). Positivity is plotted upwards. Note the clear positivity peaking at 300 ms in the antonym condition (top) as well as the negativity around 400 ms in the related (middle) and unrelated (bottom) conditions.

Figure 3

Speed-accuracy tradeoff (SAT) by condition (Experiment 2). Curve computed on grand average data. Note the lower asymptotic performance in the related condition, but otherwise similar dynamics.

Figure 4

Accuracy over time (Experiment 2). The black line represents grand average accuracy across all conditions, with the shaded region indicating the 95% bootstrapped confidence interval of the grand mean. The colored lines indicate single-subject performance.

Figure 5

Comparison of model fit. The parametric and ERP-based models yield identical fits, shown here as perfect overlap (evident in the apparent color being a mixture of the blue and red of the individual colors in the legend), and fit the overall shape of the data well. Error bars indicated 95% bootstrap confidence intervals computed on the display (response) scale.

Figure 6

Partial Effects in the ERP model. Note that the P300 predicts performance in the antonym condition and the N400 predicts performance in the unrelated condition. Neither the P300 nor the N400 predicts performance particularly well in the related condition, but the P300 seems to do a slightly better job despite the clear N400 effect and a lack of a clear P300 effect in the ERP data for the same condition. Error bars indicated 95% bootstrap confidence intervals computed on the display (response) scale.

Figure 7

Comparison of coefficient estimates with different overlap corrections for the N400. Uncertainty intervals for the frequentist models are Wald 95% intervals (i.e., twice the standard error). The uncertainty intervals for the Bayesian model is the 95% credible interval. The overall estimates are all quite close and within each other’s uncertainty intervals. The Bayesian model suggests slightly more uncertainty than the frequentist model. Note that all estimates are on the standard deviation scale.

Figure 8

Comparison of coefficient estimates with different overlap corrections for the P300. Uncertainty intervals for the frequentist models are Wald 95% intervals (i.e., twice the standard error). The uncertainty intervals for the Bayesian model is the 95% credible interval. The overall estimates are all quite close and within each other’s uncertainty intervals. The Bayesian model suggests slightly more uncertainty than the frequentist model. Note that all estimates are on the standard deviation scale.

Figure 9

Correlation of fixed effects. Each point represents a posterior sample from the Bayesian multivariate model, blue rings indicate two-dimensional density estimates. The dashed line indicates the line with unit slope through the origin, while the solid lines indicate regression lines through the samples. Strong positive correlation would show itself as the posterior densities forming ovals stretched parallel to the dashed line as well as the parallel regression lines being parallel to the dashed line, while perpendicular axes would indicate strong negative correlation. Note that neither holds: there is no strong correlation between the estimates of the coefficients for P300 and the N400.