Studies on whether word meanings are combined into sentence meanings under Continuous Flash Suppression. Experiments are based on Sklar et al. (2012). Reading and doing arithmetic nonconsciously. PNAS

We ran 4 experiments. Analysis details are in the markdown file Overall_Analysis.rmd, and an RPub containing the analyses can be found here

Experiment 1a replicates Experiment 1 of Sklar et al (2012), but with English stimuli. That experiment demonstrated lower suppression times for sentence with semantically unusual meanings. Experiment 1b is an extension in which we compare sentences to their reverse (e.g., Tom whisked the cream, The cream whisked Tom), in order to better control lexical statistics.

Experiment 2a replicates Expriment 4 of Sklar et al (2012), but with English stimuli. That experiment demonstrated that phrases with negative emotional meanings (concentration camp) break suppression faster than phrases with neutral meanings. Experiment 2b is an extension in which we compare sentences to their reverse (the baby hit the brick, the brick hit the baby). Experiment 2c is a replication of Sklar et al.'s original Hebrew experiment, to test whether visual properties of Hebrew influence participants suppression times.

## Experiment 3 This replicated our Experiments 1a, 1b, 2a and 2b, but using the original presentation scripts used in Sklar et al. (2012).

## Experiment 4 This replicated a single-word emotional CFS experiment, by Yang and Yeh (2011, Consciousness and Cognition).

The file sensicality_sims.r details a series of simulations that we ran, investigating how skewed data (of the type found in breaking continuous flash suppression experiments) can affect parametric statistical analyses such as t-tests.