Investigating the Effects of Emotion on Language Learning Using an Odor-Based Induction Method

Summary

Here, we present a protocol to measure the effects of emotional conditions on language learning, using an odor-based induction method which places participants in positive or negative emotional states by exposing them to a pleasant or unpleasant odor, and then asks them to perform a language learning task.

Abstract

Emotion has important influence on language learning. However, the effect of emotion on syntactic learning has been relatively under-researched. Most previous studies used affective pictures, music or videos to induce positive or negative emotions before having participants perform the language learning tasks. The use of these materials is an explicit induction method that might unintentionally alter participants' motivation or result in the problem of demand characteristics. To avoid such procedural artifacts, we adopted an odor-based induction approach to examining the influence of positive and negative emotions on language learning. We found that after the odor-based induction, participants in the positive-emotion group were significantly happier and those in the negative-emotion group were significantly sadder. Compared with participants in the positive emotion condition, those in the negative emotion condition performed more accurately in the assessment task, although no significant difference was found in reaction times. These findings suggest that the protocol is effective in identifying the effect of emotion on language learning. The implications of this experimental paradigm are discussed.

Introduction

Language learning is the process of learning a second language through explicit instruction and education. Emotion has profound influence on various cognitive activities such as attention, perception, reasoning, problem solving and memorization. As language learning is a process involving attention, memorization and reasoning, emotion also has a major impact on the process and outcomes of language learning¹. Most previous studies examined the relationships between learners' emotional conditions and vocabulary memorization or text comprehension². These studies intended to find out whether emotional conditions influenced the memorization or processing of emotional-congruent information. However, there have been only a few attempts to explore the effect of emotion on syntactic learning^3,4,5. How emotion influences syntactic learning is still an issue that requires further exploration and clarification.

Previous studies of the relationships between emotions and cognitive activities related to languages have adopted a wide variety of methods to place learners in a positive or negative emotional state, including affective pictures, videos, music, autobiographic recall tasks or writing tasks^6,7,8,9,10. For example, Liu et al.⁵ asked participants to listen to affective music and looked at affective pictures to induce positive and negative emotions and rated their emotional states before the language learning task was administered. In most prior studies, participants were required to evaluate affective films or pictures and rate their emotions before language tasks. However, these methods have several procedural artifacts that might affect the validity of the experimental results. First, these induction methods require the effortful processing of the affective materials¹¹. Such processing might cause participants to guess the purposes of the experiment, which might result in the problem of demand characteristics¹². In other words, once participants are aware of the experimental purposes, they might simply pretend to be in the desired emotional conditions to comply with the experimental demands¹³. Second, as the materials used in these tasks such as affective pictures, videos or writing tasks are rich in semantic information, they may interfere with the subsequent language learning tasks. For example, if a picture showing a girl with a smiley face is used to induce positive emotions, the words face, smile and girl will all be activated, and thus retrieved faster in the subsequent language learning process. If a video showing skinny children in tears is used to induce negative emotions, the words related to children, skinny, cry, eye and tears might all be pre-activated, facilitating the learning or processing of relevant information. As pictures and videos inevitably contain rich semantic information, the pre-activation of related concepts may result in biases in the outcomes of language learning. Third, the evaluation of emotional stimuli might unintentionally alter participants' motivation or alertness¹¹, which can cause some participants to perform unusually effortfully or actively. This may result in many confounding variables left uncontrolled such as arousal level, motivation or alertness, making the experimental results difficult to interpret.

In the present research, we adopted an odor-based method to induce emotions. Olfactory stimuli were used to place learners into positive or negative emotional conditions. Compared with other emotion induction approaches such as music or picture induction, the use of odors is a more implicit method to induce emotion¹⁴. Previous studies have found anatomical evidence showing that the brain structures involved in odor processing such as the amygdala and orbitofrontal cortices^15,16 are also implicated in the processing of emotional information^17,18,19,20. Negoias et al.²¹ have revealed that olfactory dysfunction was typically observed in patients with acute depression, which suggests a close connection between olfaction and emotional processing^22,23. Apart from the anatomical and pathological evidence^{14,15,16,17,18,19,20}, there is also plenty of behavioral evidence which demonstrated that odors in the environments could shape individuals' affective states in an unconscious way²⁴. Odors serve as effective affective carriers that modulate individuals' internal affective conditions¹¹. Previous studies of emotion induction showed that people exposed to pleasant odors are more prone to have positive emotions, while those exposed to unpleasant odors are more likely to be in a negative emotional condition^25,26. Odors are regarded as the ideal stimuli to manipulate the emotional state of participants largely because they are highly effective in emotion induction and they require little explicit cognitive involvement¹⁴. Therefore, odor-based induction methods enable us to investigate the effect of emotion on language learning without having to worry about the potential confounding factors such as motivation and alertness level.

So far, odor-based induction methods have rarely been used to examine the effect of emotion on language learning. Wang et al.¹¹ investigated the effect of odor-induced emotion on sentence comprehension using event related potentials. However, this study focused on the processing of participants' native language, rather than foreign language learning. Little research has been done to explore how odor-induced emotion may influence the learning of a foreign language. In this study, we intended to adopt an odor-based induction method to examine how learners in different affective states learn a foreign language differently. Compared with other induction methods, the odor-based induction method has the following three advantages. First, learners are less aware of the relationship between odor and language learning, so they are less likely to guess the purpose of the experiment. Second, as odors do not require effortful processing, participants could process them in a passive way, and thus their motivation and alertness levels are kept under control. Finally, the use of olfactory stimuli may reduce the perceptual artifacts resulting from the use of affective pictures or videos. As the olfactory channels activated by odors have no direct association with the processing of visual information, such sensory artifacts can be easily avoided, which can contribute to a more objective assessment of experimental outcomes.

Given the advantages discussed above, the odor-based induction approach might be a more effective method that enables us to identify the effect of emotion on language learning. In this study, we empirically tested the effectiveness of an odor-based induction method in the investigation into the effect of emotion on foreign language learning. The findings can also inform us of how learners in different affective states learn foreign languages differently.

Protocol

This study was approved by the Ethics Committee of Beijing Foreign Studies University and it was conducted in compliance with the guidelines for experiments with human subjects. Written informed consent was provided by all participants.

1. Preparation of olfactory stimuli

1. As this study intends to adopt an odor-based approach to emotion induction, use apple flavor to induce positive emotion and indole to induce negative emotion.
   NOTE: Prior literature showed that other smells are also applicable to emotion induction. For example, the smell of orange, strawberry or vanilla can be used to induce positive emotions, and the fish odor can be utilized to induce negative emotions²⁷.
2. Mix 200 mL of propylene glycol with 4 mL of apple flavor in a bottle to create the smell of apple juice (the pleasant smell), and mix 200 mL of propylene glycol with 5 g of indole in another bottle to create the smell of animal dung (unpleasant smell).
   NOTE: Odors that are too strong may result in the activation of the related semantic concepts. More research is needed to find out the optimal strength of odors in this type of research.
3. Ensure the bottles containing the two types of olfactory stimuli are identical.
4. Seal the bottles with stoppers to prevent odor emission.

2. Preparation of language stimuli

1. As the study focuses on syntactic learning, use experimental sentences that contain the syntactic rules of a foreign language and lexicon in participants' native language.
   NOTE: As the participants in this study were native speakers of Chinese, experimental sentences were created with Chinese words, and the word order of Japanese, a foreign language that the participants had not learned before was used. Sample experimental sentences can be found in Table 1.

Table 1: Sample experimental sentences used in the present study. (a) The experimental sentence, (b) the transliteration, and (c) its English equivalence. SOV: Subject-Object-Verb; OSV: Object-Subject-Verb; SIOV: Subject-Indirect Object-Object-Verb; S[SOV]V: Subject[Subject-Object-Verb]Verb. The sentences were adapted from our previous study⁵.

2. Include multiple word-order rules to ensure the comprehensive assessment of learning outcomes. Design at least 30 experimental sentences for each word-order rule, among which 20 sentences will be used in the training phase and 10 sentences will be used in the testing phase.
   NOTE: As shown in Table 1, four types of structures were used in this study, including such structures as S[SOV]V and SIOV. A total of 128 sentences were developed, with 80 sentences used in the learning phase and 48 used in the testing phase.
3. Make sure there is no semantic relation between the two odors and the language stimuli. For example, avoid using the odor-related words such as smell, apple or odor in the experimental sentences.
4. Randomize the sentences before presenting them to participants visually on the computer screen. To do this, select the Selection tab of the Property Page in the stimulus presentation software (Table of Materials) and set the selection method to Random.

3. Participant recruitment and preparation for the experiment

1. Recruit participants who have no background of the foreign language that will be tested. Ensure participants have self-reported normal sense of smell and normal (or corrected-to-normal) vision.
2. Assign the participants into two groups, with each group containing at least 30 members. Make sure the two groups do not differ in years of education and gender ratio. Exclude the participants who report any history of dyslexia.
3. Inform the participants that they should be free from exhaustion, hunger, illness or other conditions that make them uncomfortable on the day of the experiment.
4. Invite the participants in groups or individually to the laboratory.

4. Procedure

1. Invite subjects to the laboratory room and instruct them to sit down at the computer workplaces.
2. Present the written informed consent forms to subjects. Instruct them to read and sign the forms.
3. Present participants the pencil-and-paper version of the Self-Assessment Manikin (SAM) pictorial rating scale^28,29, and ask them to rate their baseline emotional conditions in terms of valence, arousal and dominance by marking with a pen.
4. Expose participants to the olfactory stimuli, using odor dispensers. Ask participants to familiarize themselves with the odor for 10 min. Make sure the length of exposure is the same for the positive-emotion group and the negative-emotion group.
5. Ask participants to rate their emotional conditions again using the SAM scale.
6. Ask participants to perform the training task.
   1. Ask participants to read the instructions visually presented on the computer screens: "You will learn a new language which contains Chinese words and a new grammar. Next you will see some sentences in this language. Please observe them carefully and try to learn their grammatical structures. Press any key to start when you're ready."
   2. Present experimental sentences on the computer screens and have the participants observe the sentences for syntactic regularities.
   3. Present the following written instructions on the computer screen and ask participants to read them carefully: "Next you will see more sentences. Please read them carefully and decide whether they are grammatically correct. Press '1' for grammatically correct sentences and press '0' for incorrect sentences. You will see the correct sentences after your responses. Press any key to proceed to the experimental task."
   4. Ask the participants to judge the grammatical acceptability of the sentence on the computer screen by pressing a button ('1' for grammatical and '0' for ungrammatical) and present feedbacks ('CORRECT!' or 'INCORRECT!') on the screen after each response. Present the correct structures after the feedbacks to reinforce the effect of learning.
   5. Present the following written message on the computer screen to inform participants that the task is completed: "The learning task is completed!"
7. Instruct participants to rate their affective states with the SAM scale after they finish the learning task.
8. Ask participants to perform the testing task.
   1. Present the following written instructions on the computer screen for participants to read: "Next you will see more sentences. Please decide whether they are grammatically correct. Press '1' for grammatically correct sentences and press '0' for incorrect sentences. You will have five seconds to respond to each question. Please try to respond as accurately and as quickly as you can. Press any key to start the experiment."
   2. Ask the participants to judge the grammatical acceptability of the sentence with a button press ('1' for grammatical and '0' for ungrammatical).
   3. Present the following message on the screen to inform participants that the experiment is over: "This is the end of the experiment. Thank you for your participation!"
9. Ask the participants to fill in the questionnaires about their demographic details such as age, education and gender, and answer a question concerning the possible purpose of the experiment.
10. Provide participants monetary compensation or rewards for their participation in the experiment.

Results

The results of emotion ratings are summarized in Figure 1. Repeated-measures ANOVA was performed with valence ratings as the dependent variable, and group (positive, negative) and time (before induction, immediately after induction, after learning) as the independent variables. The results showed a significant effect of group, F (1, 58) = 24.71, p < 0.05, and a significant interaction effect between group and time, F (1, 58) = 2...

Discussion

This study investigated the effect of emotion on the learning of syntactic rules in a foreign language, using an odor-based emotion induction approach. We tested the learning performance in a positive-emotion group and a negative-emotion group. Participants were first exposed to a pleasant odor or an unpleasant odor. Then they were instructed to learn the syntactic rules of a foreign language. Finally, a grammaticality judgment task was administered to assess their learning outcomes. The critical step within this protoco...

Materials

Name	Company	Catalog Number	Comments
Apple flavor	Givaudan	N/A	Used to induce positive emotion
Computers	N/A	N/A	Used to present stimuli and record subjects' responses.
E-prime	PST	2.0.8.22	Stimulus presentation software
Indole	Taida	N/A	Used to induce negative emotion
Self-Assessment Manikin (SAM)	N/A	N/A	Used to assess subjects' affective states. From Lang (1980).