Ferreira and Pashler found that pictures with high frequency names and pictures preceded by high constraining sentences were named faster than pictures with low frequency names and pictures preceded by weakly constraining sentences. They also found that the tone discrimination latencies were affected by these semantic and morphological manipulations. The longer the semantic and morphological processes took, the later participants responded to the tone. This suggests that concurrent processing cannot occur during lemma selection or during morpheme retrieval, a finding that converges with observations of eye movements during lexical access reported by Meyer et al (1998) and Meyer & van der Meulen (2000).
In Experiment 2, Ferreira and Pashler manipulated phoneme retrieval by presenting the picture along with a written distracter word that was phonologically related or unrelated to the picture name. As is usually the case, a priming effect for speech onset latencies was found (e.g., Schriefers et al, 1990). However, tone discrimination latencies were not affected by priming suggesting that phoneme selection is not a demanding process. In contrast, eye tracking studies have demonstrated that viewing times are influenced by phonological variables such as word length and priming which are located at the phoneme selection level (Meyer & van der Meulen, 2000; Meyer, 1996; Meyer & Schriefers, 1991).
In one such eye tracking study, participants were presented with a picture-word interference task in which they had to name two pictures per trial (Meyer & Van der Meulen, 2000). Simultaneously the participants heard distracter words that were either phonetically related or phonologically unrelated to the left object’s name. They found that both reaction and viewing times before naming objects were shorter after phonologically related distracters than after phonologically unrelated distracters. This suggests that speakers look at, and attend to, the objects they name until they have retrieved the phonological forms of their names. As noted above, a plausible reason for this is that the phonological representation is the representation that is accessed in capacity demanding self-monitoring processes. In contrast, Ferreira and Pashler’s results suggest that the generation and, presumably, further use of the phonological representation does not require central resources.
Complete phonological and at least partial phonetic preparation of a name before a shift of gaze was reported by Meyer, Roelofs and Levelt (2003). In their fourth experiment the speakers reaction and viewing times were measured while they were naming object pairs in noun phrase conjunctions such as ‘peer and schaa’ (pear and scissors). The word length of the left objects name was manipulated so that half of the target objects had monosyllabic names and the other half had disyllabic names. The experiment had two conditions: a mixed block (where half the object names were monosyllabic and half disyllabic) and a pure block (where all object names had the same length). They found that in the pure block only, reaction and viewing times were shorter when the left objects name was monosyllabic than disyllabic.
According to Meyer et al. (2003) the reason why a word length effect was only observed in the pure blocks is because speakers use a different criterion in deciding when to speak for monosyllabic and disyllabic object names. When speakers are presented with the objects to name in pure blocks they aspect each trial to be of similar difficulty and so use an optimal criteria. In contrast, in mixed blocks each trial varies in difficulty so speakers use a criterion that is acceptable for both monosyllabic and disyllabic names, but not optimal for either. As the length effect in the pure condition was reflected in the speakers viewing times it can be assumed that an object’s name is planned on the phonological and at least to a certain extent on the articulatory level before a gaze shift is initiated.
In the experiment reported below, Ferreira and Pashler’s Experiment 1 of single-object naming was replicated. It was expected that the same pattern of results as in Ferreira and Pashler’s study would be found, i.e. a semantic but no phonological effect on the tone-monitoring latencies. In order to explain the discrepancy between the results obtained by Ferreira and Pashler and in the eye tracking studies a second experiment is planned. In this second experiment the reaction time and eye movement studies will be combined so that speakers will now be required to name two objects per trial (as in the eye tracking experiments) and react to the tones (as in Ferreira and Pashler’s experiments). This should also contribute to our understanding of why speakers look at objects they name for as long as they do.
Finally, a third experiment is planned whereby speakers will only name the object pairs (there will be no tone discrimination task), but their eye movements will be recorded. In Experiments 2 and 3 it is expected that compatible results will be found for the tone monitoring latencies and the inspection times for the objects. Based on the earlier eye tracking experiments, it is expected that both variables will show effects of semantic and phonological difficulty. The expected discrepancy to Experiment 1 might be due to the fact that the phonological encoding or the monitoring processes may be much easier in single word than in phrase production.
Experiment 1
Method.
Participants. The experiment was carried out with 11 students of the University of Birmingham, whose native language was English. They had normal or corrected-to-normal vision and received course credits or money for their participation.
Materials. The experimental pictures were 32 line drawings of common objects selected from the picture database of the Max Plank Institute for Psycholinguistics and the Snodgrass and Vanderwart (1980) collection. (see Appendix A). The experiment had four conditions: conceptually related, conceptually unrelated, phonologically related, and phonologically unrelated. The words in the conceptually related condition were categorically related to the picture names (e.g., “sheep-goat”). In the conceptually unrelated condition, the pictures were recombined with the distracter words from the other conceptually related pairs. The words in the phonologically related condition shared word-initial segments with the picture names (e.g., “elephant-element”). Finally, in the phonologically unrelated condition, the pictures were recombined with the distracter words from the other phonologically related pairs.
Design. The experiment consisted of four test blocks, in each of which the 32 experimental items were shown once preceded by four practice items, which were the same for each block. The experimental design included two independent within-subject variables: distracter relatedness (conceptually related, conceptually unrelated, phonologically related, or phonologically unrelated) and tone SOA (50, 150, or 900 ms). The order of the items within each block was random and different for each block and participant. The dependent measures were picture-naming and tone discrimination latency.
Apparatus. The experiment was controlled by the NESU 5.01 program of the Max Planck Institute for Psycholinguistics, Nijmegen. The pictures were presented on a Samtron 95 Plus 19 inch screen. Speech was recorded using a Sony ECM-MS907 microphone, a Sony TCD-D8 DAT recorder and was recorded on the hard drive of a Linux machine. Auditory stimuli were presented through a pair of Beyerdynamic DT 931 headphones. Button press responses were measured using two response boxes with two push buttons on each, of which one push button was masked.
Procedure. Participants were tested individually and were seated in a sound-attenuated booth. At the beginning of the experiment, the participants received a booklet showing the pictures used in the experiment and were asked to name them aloud. Naming errors were corrected. After interactive instructions (see Appendix B), participants proceeded through three practice phrases. In the first phase, participants were given practice at tone discrimination alone for 45 trials (15 trials of each pitch). In the second and third phrases participants saw the 32 experimental pictures to name and heard the tones to discriminate. In the second phase, no distracters were presented, whereas in the third, a new unrelated distracter was presented simultaneously with the picture.
At the beginning of each trial, a fixation point was presented in the centre of the screen for 1,000 ms. Following a blank interval of 500 ms, the picture stimulus was presented and remained on the screen until the voice key detected a response. The distracter word was presented in bold, Courier 16-point font simultaneously with the picture and remained on the screen for 200 ms. Participants were instructed to name the picture as quickly and as accurately as possible and to ignore the visual distracter word. Order of presentation was random, with the provision that no more than two items in the same condition would appear consecutively.
The auditory stimulus for tone discrimination was presented 50, 150, or 900 ms after picture onset. The tone was 285 ms in duration and was either low (180 Hz), medium (500 Hz), or high (1200 Hz) in pitch. The pitch of the tone that participants heard varied randomly between trials. Participants were instructed to name the picture promptly, while still identifying the pitch of the tone as quickly as possible. The response buttons were labelled “low,” “medium,” and “high” from left to right, which participants pressed with the index finger of the left hand and the index and middle fingers of the right hand, respectively. Each trial ended when both a voice key response and a button-press response were registered. After a pause of 500 ms the next trial began.
Results.
The data from three participants was excluded due to an error occurring on the NESU program, one participant been unable to come back for the second session and because one participant performed very poorly. The data from 54 trials (0.05%) in session 1 and 27 trials (0.03%) in session 2 were discarded because participants used incorrect object names or responded with picture-naming latencies of more than 2000 ms.
Table 1
Mean naming latencies by distracter relatedness for session 1 and session 2
Task 1 performance
Picture-naming latencies were significantly faster in session 1 compared to session 2 (F 1, 7) = 14.07, P < 0.007). As table 1 shows, in session 2 only, conceptually related distracters significantly slowed picture naming latencies compared to conceptually unrelated distracters (F 1, 7) = 8.42, P < 0.023). No other effects approached significance for the picture naming response latencies.
Task 2 performance
In addition to the data excluded from the three participants mentioned earlier, the data from one other participant was excluded from the tone discrimination analyses due to them performing significantly worse on the tone discrimination task in session 2 compared to session 1. The tone discrimination latencies for the remaining 7 participants were significantly faster in session 2 than in session 1 (F 1, 5) = 4.20, P < 0.096).
A PRP effect was observed in both sessions. In session 1, tone discrimination responses were 228 ms slower when tone onset occurred 50 ms after picture onset compared to when it occurred 900 ms after (the difference between the 50 ms and 150 ms SOA conditions was 54 ms). This is supported by a significant main effect of tone SOA (F 2, 12) = 18.88, P < 0.000). In session 2, tone discrimination responses were 162 ms slower when tone onset occurred 50 ms after picture onset compared to when it occurred 900 ms after (the difference between the 50 ms and 150 ms SOA conditions was 32 ms). This is supported by a significant main effect of tone SOA (F 2, 12) = 10.20, P < 0.003).
The main effect of distracter relatedness was significant only in session 1 (F 3, 18) = 3.53, P < 0.036). Conceptual related distracters significantly slowed the speed of tone discrimination responses (compared to conceptually unrelated distracters) in session 1, but only in the 50 ms condition (t = 2.348, df = 6, P < 0.057).
Discussion .
The results presented here support the PRP effect: as tone SOA decreased, tone discrimination responses increased. However, the other results obtained were not as expected. From Ferreira and Pashler’s study it was expected that tone discrimination responses would be slowed by conceptually related distracters as much as picture naming latencies were. This was not observed in the present study. In session 1, conceptually related distracters did not affect the speed of picture naming latencies; however tone discrimination latencies were slowed by this conceptual manipulation. The opposite pattern of results was observed in session 2, in which picture naming latencies were slowed by conceptually related distracters but tone discrimination latencies were not. The phoneme manipulation also did not follow the expected pattern of results. Ferreira and Pashler reported that phonologically related distracters facilitated picture naming latencies but did not affect the speed of tone discrimination. In the present study the phoneme manipulation did not produce any significant results.
The analyses and results of the present experiment were weakened due to faults with the design of the experiment. This experiment will be carried out again with the following necessary improvements in Placement 2. The main problem was that the distribution of the tone SOAs and beeps were unevenly distributed within each of the four blocks due to human error. This will need to be corrected in Placement 2.
One difference between the present study and Ferreira and Pashler’s was their use of a mask of Xs to replace the distracter word after 200 ms. In the present study, instead of a mask of Xs replacing the distracter word, the word disappeared after 200 ms. Their reason for masking the word is still awaiting a reply. A second difference between the two studies is that in the present study participants were required to carry out the experiment on two separate occasions whereas in Ferreira and Pashler’s, participants were only required to carry out the experiment once. Carrying out the experiment seemed difficult due to its dual task design so participants in the present study were asked to carry out the experiment twice to see how practice would affect the results. All participants except one (who performed worse on the tone discrimination task in session 2 compared to session 1) were significantly faster at picture naming and tone discrimination in session 2 than session 1. In Placement 2 extra practice blocks will be given at the start of the experiment.
Since no response box with three buttons could be found in the department (and ordering one would take too long) two dual response boxes were used in the present study. Analysing the tone discrimination latencies from the two response boxes proved to be confusing and provided extra work. At present the availability of a response box with three buttons together with the necessary software is been looked into. Similarly, confusion and extra work was made by using a Linux machine to analyse the speech latencies. Various microphones and headsets were tested, but when pressing the buttons for the tone discrimination task each microphone was triggered. The only alternative in the short-term was to use the Linux machine. To cut out this unnecessary confusion and extra work a reply from Ferreira is awaiting to see what apparatus he used.
The power of the results reported in the present study is further weakened by only carrying out the experiment with nine participants, whereas Ferreira and Pashler used forty-eight. In Placement 2 more participants will be recruited. The experience of the problems mentioned above has provided me with a better understanding of the design of the experiment to implement the necessary improvements.
References.
Ferreira, V.S., & Pashler, H. (2002). Central bottleneck influences on the processing stages of word production. Journal of Experimental Psychology: Learning, Memory, and Cognition, 28, 1187-1199.
Griffin, Z.M. (2001). Gaze durations during speech reflect word selection and phonological encoding. Cognition, 82, B1-B14.
Griffin, Z.M., & Bock, K. (2000). What the eyes say about speaking. Psychological Science, 11, 274-279.
Jescheniak, J. D., & Levelt, W. J. M. (1994). Word frequency effects in speech production: Retrieval of syntactic information and of phonological form. Journal of Experimental Psychology: Learning, Memory and Cognition, 20, 824-843.
Levelt, W.J.M. (1983). Monitoring and self-repair in speech. Cognition, 14, 41-104.
Levelt, W.J.M. (1989). Speaking: From intention to articulation. Cambridge: MIT Press.
Meyer, A.S., & Schriefers, H. (1991). Phonological facilitation in picture-word interference experiments: Effects of stimulus onset asynchrony and types of interfering stimuli. Journal of Experimental Psychology: Learning, Memory, and Cognition, 17, 1146-1160.
Meyer, A.S. (1996). Lexical access in phrase and sentence production: results from picture-word interference experiments. Journal of Memory and Language, 35, 477-496.
Psychological Review, 104, 3-65.
Meyer, A.S., Sleiderink, A.M., & Levelt, W.J.M. (1998). Viewing and naming objects: Eye movements during noun phrase production. Cognition, 66, B25-B33.
Meyer, A.S., & Meulen, F.F. van der. (2000). Phonological priming of picture viewing and picture naming. Psychonomic Bulletin & Review, 7, 314-319.
Meyer, A.S., Roelofs, A., & Levelt, W.J.M. (2003). Word length effects in object naming. The role of a response criterion. Journal of Memory and Language. 48, 131-147.
Schriefers, H., Meyer, A.S., & Levelt, W.J.M. (1990). Exploring the time course of lexical access in language production: Picture-word interference studies. Journal of Memory and Language, 29, 86-102.
Pashler, H. (1994). Overlapping mental operations in serial performance with preview. Quarterly Journal of Experimental Psychology, 47A, 161-191.
Pashler, H. (1998). . Philadelphia: Taylor & Francis Press.
Wheeldon, L., & Levelt, W.J.M. (1995). Monitoring the time course of phonological encoding. Journal of Memory and Language, 34, 311-334.
Appendix A.
The material used in the experiment.
Appendix B.
Instructions that were given and explained where necessary to participants.
You will be given three practice blocks:
1st block: you will hear one of three tones: low, medium or high. Your task is to identify the tones.
low =press the button labelled low with the index finger of your left hand.
medium =press the button labelled medium with the index finger of your right hand.
high =press the button labelled high with the middle finger of your right hand.
Press the buttons as quickly and as accurately as possible.
2nd block: 36 practice pictures will appear on the screen in front of you plus a tone 50, 150 or 900 ms after picture onset. Your task is to name the picture as quickly as possible while promptly identifying the tone as low, medium or high with a button press.
3rd block: the 32 experimental pictures will appear on the screen in front of you but this time with a distracter word over them plus a tone 50, 150 or 900 ms after picture onset. Your task is to name the picture (and ignore the distracter word) as quickly as possible while promptly identifying the tone as low, medium or high with a button press.
The experimental stimuli will be presented in four blocks in the same format as practice block 3. At the beginning of each experimental block, 4 practice pictures with practice distracter words will appear on the screen in front of you. You will be given a short break between each block.
If you have any questions, please ask now.
Thank you!