Native Language Experience Shapes Pre-attentive Foreign Tone Processing And Guides Rapid Memory Trace Build-up: An ERP Study Part 3

1.3 | The present study

The above-outlined literature has provided important insights into the processing of L2 tone. Importantly, it has shown a reliance on phonetic knowledge for higher-level tone learning to occur, such as the mapping between tone and lexical or grammatical content. 

Grammar is the basis of language. It is an important tool for us to express ourselves and transmit information. Good grammar not only helps us express our thoughts more accurately and fluently, but also improves our communication skills, enhances self-confidence, and enhances our image. Therefore, grammatical content and memory are closely related.

Learning grammar well requires us to constantly repeat and memorize it, and master the rules and applications of various language structures. This places strict demands on our memory and requires us to maintain a patient and perseverant attitude in the long-term learning process. If we can use memory skills in the learning process, such as expanded memory, associative memory, etc., we can master grammatical knowledge faster and improve memory efficiency.

At the same time, good grammar skills can also help us better understand and remember other knowledge. For example, subjects such as science and history are often accompanied by a large number of professional terms and articles with complex sentence structures. Good grammar skills can help us understand these contents more quickly and accurately, and deepen our memory.

Therefore, we should realize that grammatical content and memory are closely related. A good grammar level can help us improve our memory, thereby continuously improving our learning and expression abilities. We should conduct regular language training and constantly memorize and consolidate words to achieve better language expression skills. It can be seen that we need to improve memory, and Cistanche deserticola can significantly improve memory because Cistanche deserticola is a traditional Chinese medicinal material that has many unique effects, one of which is to improve memory. The efficacy of minced meat comes from the various active ingredients it contains, including acid, polysaccharides, flavonoids, etc. These ingredients can promote brain health in various ways.

Click know supplements to boost memory

The acquisition of grammatical tone is likely easier than that of lexical tone as grammatical function is mapped directly onto the phonological tone categories rather than the tones being meaningful in association with segmental information only. 

Due to their previous phonetic sensitivity and phonological experience with tones, lexical tone speakers may be able to acquire grammatical tone more easily than grammatical tone speakers do lexical tone. 

In general, previous research has suggested that L1 familiarity with tone or other previous tone experiences can facilitate the discrimination of L2 tone or the acquisition of novel tonal words. However, it is largely unknown what degree of familiarity or similarity is needed for facilitation to occur, and if transfer facilitates grammatical tone learning, too. 

It is also unclear whether L1–L2-based facilitation affects all levels of tone processing alike. To address some of these issues, we studied two groups of learners during their acquisition of artificial novel words with grammatical tone. 

All learners came from closely related and highly similar languages, but one group's native language was tonal (Swedish), and the other's was not (German). We manipulated the similarity of L1 and L2 tone for the tonal learners (contour tones [fall, rise] vs level tones [high, low]) and investigated different types of learning responses: the pre-attentive gating response for auditory stimuli, late responses related to lexical and grammatical processing, and behavioral responses for mismatch detection. Further, because we chose a grammatical type of tone, tones could be studied independently of the lexical items they were attached to. 

This also allows us to investigate the importance of the learners' familiarity with the tone's linguistic function directly-in this case, the expression of grammatical meaning. It is possible, for instance, that we find amplitude differences in the N400/AN for all novel words with native-like tone patterns in the early acquisition stages as a form of direct transfer and L2 processing through the lens of the L1 (e.g., a reduced N400 for easier lexicosemantic processing compared with pseudowords or an increased AN for rule-based processing after successful rule acquisition). 

In the behavioral data, investigating tone mismatch identification, we could study the behavioral correlates of tone acquisition fully isolated from the segmental information. Any familiarity-based advantage observed at this point could be directly attributed to L1– L2 familiarity. 

As the grammatical tones that we used involved a one-to-one mapping with meaning, likely, phonetic attunement and categorization into phonologically relevant units would directly and immediately open up for the acquisition of the tone's grammatical content. 

We assumed that if grammatical function and phonological similarity of L1 and L2 tones would be the determining factor in L2 tone acquisition, the tonal speakers' learned attention to pitch movement should result in facilitated tone-grammar learning (faster word trace formation and changes in N400/AN3 ) for words with tonal movements, possibly even restricted to falling pitch, as the only tone that has word-level relevance in their L1. 

If, however, general experience with grammatical tone is sufficient to facilitate L2 tone acquisition, we should see no differences between tone types for the tonal L1 speakers but still clear differences between tonal and nontonal learners. 

If L1 experience should not influence the discrimination of tones and the mapping between tone and grammar in its initial stages, there should be no processing differences between learners from different L1 backgrounds. 

Finally, the different responses that we investigated could be affected differently by L1– L2 familiarity. 

It has previously been shown that transfer affects early processing stages more than behavioral responses and late processing stages (Andersson et al., 2019). Therefore, we expected the strongest effect of familiarity in the early response.

2 | METHOD

2.1 | Participants

Forty-eight healthy, right-handed adults (mean age 23.7, 25 females) with normal or corrected-to-normal vision and normal hearing (defined as pure-tone hearing thresholds ≤20 dB Hearing Level (ISO, 2004) were recruited for the study. 

All tests were carried out at the Lund University Humanities Lab, and most participants were students at Lund University. Half of the participants had a tonal L1, Swedish, and the other half a nontonal L1, German. 

Twenty-four participants were chosen per group to allow for the counterbalancing of vowels and tones between groups; see below. The tonal and nontonal participants were each divided into two learner groups: high/fall learners (i.e., participants who were taught target words with high and falling tones, where low and rising tones served as controls) and low/rise learners (i.e., participants who were taught target words with low and rising tones, where high and falling tones served as controls). 

The division into high/fall vs low/rise was based on the desired property of all target words to initially have identical pitch and be indistinguishable from each before the onset of the vowel, which was also the onset of the tone movement. 

In this way, we obtained a clear divergence point for the ERP data. All four groups (i.e., tonal L1 high/fall, tonal L1 low/rise, nontonal L1 high/fall, nontonal L1 low/rise) were matched for age (23-24 years), gender (6 females per group), socioeconomic status (Hollingshead, 1975), working memory span (Unsworth et al., 2005), their perception of nontonal phonological contrasts (i.e., vowel duration: mean accuracy 97.1%) and their discrimination of extralinguistic pitch (i.e., piano tones: mean accuracy 92.5%). All subjects were remunerated for their participation. 

One tonal L1 high/fall participant reported previous exposure to a foreign tone language and one participant from the nontonal L1 low/rise group chose to discontinue the experiment. The data from both participants were excluded. The experiment was carried out according to the guidelines in the Declaration of Helsinki and approved by the local ethics review board in Lund.

2.2 | Stimuli

To test tone processing and tone word acquisition, we created 24 tonal pseudowords. We chose an artificial language instead of a natural grammatical tone language to control the experimental situation. Artificial language learning has been shown to correlate strongly with natural language learning and was, therefore, deemed suitable for this experiment (cf. Ettlinger et al., 2016). 

All test words had a simple CVC structure (consonant-vowel-consonant), for instance,/siːs/, a frequent structure for monosyllables in German and English. Monosyllables were deemed suitable for studying transfer from Swedish as the domain of tone in Swedish is the word rather than the syllable (in contrast to, e.g., Mandarin) and because monosyllabic stems frequently and most consistently undergo tone changes. 

All consonants and vowels were recorded separately in an anechoic chamber by a male speaker of Russian (to prevent a bias that would arise should the speaker come from one of the two experimental L1s). 

During the recording, consonants were preceded or followed by two dummy vowels(/o/,/ø/) for their naturalistic pronunciation unconfounded by coarticulation with some of the actual stimuli's vowels. 

These dummy vowels were cut off before splicing the consonants with the actual vowels used for the main test words (/a/, /ε/, /i/, /u/). The initial consonants, vowels, and final consonants were then equalized for length and loudness and spliced together in Praat (Boersma, 2001) with 10 ms transition phases. All resulting pseudowords were 1000 ms long (C = 328 ms, V = 464 ms, C = 218 ms) with short silent closures before an initial and after a final plosive. 

The employed Russian phonemes, while similar to both German and Swedish ones, were chosen to avoid differential carry-over effects that German or Swedish phonemes could have evoked. All test words were perceived equally well and correctly classified as pseudowords by eight German speakers and three Swedish speakers who were not participants in the main study. 

In the final step, we used pitch manipulations to add two level tones (high: 138 Hz and low: 98 Hz) and two contour tones, a rise (98 Hz to 138 Hz) and a fall (138 Hz to 98 Hz). The pitch was selected by the speaker's natural pitch range. 

The pitch movements had a naturalistic pitch span (40 Hz, 6 semitones) compared with the pitch movements of Swedish word tones and were easily distinguished by several native speakers of German and Swedish who piloted the experiment. 

The onset of pitch movement was aligned with the onset of the vowel (cf. Figure 2) to define a single earliest possible point at which the stimuli could be identified. This point was used as a time-locking point for the ERP data. 

Yet, while it is the earliest point at which the stimuli diverge, it would take a few milliseconds for the listeners to distinguish and correctly identify the stimuli. This presumably varied slightly, both intra- and interpersonally.

For more information:1950477648nn@gmail.com