3 thoughts on “Experience and conceptual overlap modulate cross-language priming effects

  1. Thank you very much for your very interesting studies.
    Q1) I was just wondering, in Table 2, why the results for related concrete words were not very different between L1 to L2 and L2 and L1.
    The significant effect seems to be due to the slower RTs for unrelated in L2 to L1 than in L1 to L2.

    Q2) Figure 2: What are predicted inverse-transformed RTs? Sorry for a naive question.


  2. Dear Yoko,

    Thank you very much for your interest in the study.
    A1) You are right in that the RTs do not differ significantly when comparing the words in the Related condition. The priming asymmetry between priming effects in the L1-L2 vs L2-L1 directions for concrete word pairs is, as you have noticed, driven by the Control primes. At first sight, that would suggest an inhibitory effect of those Control primes on the processing of the targets (i.e., participants took more time to process the target when they needed to reject the information of the unrelated prime). Alternatively, it could also be explained by a similar benefit from the Related primes in both translation directions. In that case, the different RTs in the Control condition would be simply due to participants responding more slowly to the L2 targets than to the L1 ones (i.e., without any inhibitory effect).

    At the moment, I am not sure which can be the most suitable explanation. However, after inspecting the data, two things make me consider the second option as more plausible. (1) When I examine the prime frequency effect, it seems to suggest that there is only benefit from the Related primes (and not inhibition from the Control ones). (2) There is a general effect of Language: Participants do respond more slowly to L2 targets (independently from frequency—but not from L2 use).

    Without going any further, I would say that I “prefer” the second explanation from the two exposed at the beginning with respect to your question. However, that leaves me with a bigger problem, that is, why are the RTs for Related primes so similar? I am afraid I do not have a straight answer for this at the moment. Besides, we should also consider that those are the general priming effects, that is, without considering frequencies and L2 use; thus, they might be quite misleading.

    A2) Response times where inverse-transformed so that the assumptions of the linear mixed-effects models were met. That figure just shows the predicted (inverse-transformed) response times from the model. (For the priming effects, I just showed the standard response times, which were in milliseconds, by the way.)

    I hope I answered your questions. Please, let me know if you need any other clarification. You can always send me an email to a.chaouchorozco@pgr.reading.ac.uk.

    Thank you again for your message.


Leave a Comment or Question Below