The Effects of Pre- versus Post-Presentation Input Flooding via Reading on the Young Iranian EFL Learners’ Acquisition of Simple Past Tense

The purpose of this study was to find out the possible effects of pre-versus post-presentation input flooding via reading on simple past tense acquisition among young Iranian EFL learners. Sixty one elementary EFL learners were divided into two experimental and one control group. The experimental groups were exposed to pre and post presentation input flooding via reading while the control group received the traditional method of teaching grammar at a private language institute. A dictogloss activity was also used to check the number of ongoing grammatical errors in all groups. The treatment was conducted in thirty minutes of twelve sessions for about two months. ANCOVA and one way ANOVA results of both post-test and dictogloss indicated the significant effects of preand post-presentation input flooding via reading on the acquisition of simple past tense. However, the results of posttest indicated that preand post-presentation input flooding groups performed fairly similarly while the results of dictogloss showed the outperformance of the former to the latter.


Introduction
As the grammar instruction is an imperative matter to be considered in Iranian classes, instinct and information are examined to be considered as the treatments which can enhance the issues confronted in the classroom.Likewise, to the best of the researcher's information, there was no study found in the literature to research the timing of data flooding.Also, out of the researcher's experience and checking and other instructors' report, simple past tense is respected to be a risky zone in L2 obtaining.
Input is a paramount matter in L2 teaching and as Ellis (1998) expressed, acquisition happens when learners go to the new structure in data as opposed to when they deliver it.This implies that the learners must notice and pay attention to the new aspect of language in order to make the information part of their interlanguage.According to Sharwood Smith (1991), one way of dealing with the matter of input is to enhance the noticeability of it through increasing its perceptual salience.Moreover, Rutherford and Sharwood Smith (1985) stated that this enhancing of noticeability is termed as consciousness raising and later as input enhancement.Gass and Selinker (2008) expressed that if a component happens often in the data, it is liable to be taken note; hence, this study will investigate the role of input flood in learning English grammar.
Input flood, as one of the forms of input enhancement, which is the focus of this study is a technique in which the learners are provided with huge amount of examples of a certain target form in the input that can take the form of either oral or written.Its assumption is that frequent occurrence of the same target form makes it more noticeable and it draws the learner's attention to the intended form (Fotos & Nassaji, 2011).
According to its importance and significance, input matters in grammar instruction are of the crucial challenges on the part of the teachers.Therefore, different techniques such as input enhancement, textual enhancement and input flood are proposed to deal with this problem.There are different options for teachers to realize input enhancement.Teachers may provide the students with texts of bold type to promote noticing (textual enhancement) or they may flood the target structure through different materials useful in learning the intended grammatical point.They may also figure out that these techniques should be followed by direct presentation of the target structure or they may suffice by themselves.Ellis (2001) characterizes Form Focused Instruction (FFI) as "any planned or incidental instructional activity that is intended to induce language learners to pay attention to linguistic form" (p.2).According to Ollerhead and Oosthuizen (2005), FFI is utilized as a general term for distinctive ideas, for example, 'logical instructing', 'Focus on Form (FonF)', Focus on FormS (FonFS)', 'remedial input/ slip revision' and 'transaction of structure'.FonF is referred to as a methodology to L2 showing where thoughtfulness regarding structure emerges from exercises that are essentially significantly concentrated (Long & Robinson, 1998).

Review of Related Literature
Noticing is defined by Schmidt (1990) as "a hypothesis that input does not become intake for language learning unless it is noticed, that is, consciously registered" (p.130) and therefore is seen as a pre-requisite for language learning process.It is also defined by Fotos (1993) as "the mechanism by which learners, after sensitization to a particular structure, "spot" such structure (or its absence) in subsequent natural input" (p.383).
As defined by Leow (2001), enhanced input is a term used for (mostly written) input that has been altered typologically to enhance the saliency of target forms.Such typological means may include italicization, bold print, underlining, shading, the use of different font types and sizes, and capitalization.
The grammar instruction is defined by Schmidt and Frota (1986) as constituting any instructional technique to help students notice the gap between learners' interlanguage and the new features in target language structure.This is the process which leads to grammar discovery and hypothesis construction (ibid).The case of noticing in this process is of a great importance and it leads to fast learning of the intended grammatical point.
Input flooding is defined by Sharwood Smith (1993) as the enrichment of input by supplying numerous examples of the target form without overtly drawing attention to it.It provides the learners with ample exposure to the target form.Because of the avoidance in manipulation of any kind of intervention in this technique, it is considered as an implicit method of focus on form.An example of the flooded input via texts which is intended for the use of past tense will follow: A chipmunk sat on some branches in a great big tree.It was very hungry, so it decided to leave the tree and look for food.It climbed off the branches and reached the trunk of the tree, and went down the truck to the ground below.The chipmunk saw lots of grass, and in the grass lay many acorns!The chipmunk, in its delight, took as many acorns as it could, put them in its mouth, and ran back up the tree trunk to its nest.There, the chipmunk had a very good meal (Fotos & Nassaji, 2011, p. 43).Scott and De La Fuente (2008) led a qualitative study to research the part of first language (L1) in awareness raising and structure centered linguistic use assignments.For this reason, moderate level school learners of French and Spanish learners were occupied with the study.The utilization of students' L1 and L2 to solve grammar problems were explored using conversation analysis audio-taped interactions and stimulated recall sessions.The ones who were permitted to utilize L1, worked synergistically as a part of an adjusted and rational way while the ones in gathering two were obliged to utilize the L2 and showed divided cooperation and little proof of coordinated effort.Discoveries from animated review sessions proposed that perusing, thinking and talking gave off an impression of being synchronous and incorporated methods for the understudies in group one; nonetheless, these methodologies seemed, by all accounts, to be consecutive and looking at for the members of group two.These discoveries likewise welcome the educators to adapt to the issue of the L1 in the outside language classroom.Trahey and White (1993) analyzed whether input flood was enough to empower the French learners of English as a L2 to discover that English allows the insertion of a modifier between the subject and the verb (French does not) yet does not allow to position it between the verb and the article (French does).The learners were presented to the data one hour a day for 10 days.The target structure was not over emphasized amid the process.The learners succeeded in taking in the subject-intensifier-verb (SAV) position yet neglected to "unlearn" the ungrammatical subject-verb-modifier-object (SVAO) position which was allowed in their first language.In a deferred post-test managed one year after the treatment, nonetheless, Trahey (1996) found that the helpful impacts of input flood on the procurement of SAV had vanished.Along these lines, this study proposed that the impacts of enriched input on procurement are constrained and it is not enough by itself.
Later, Loewen, Erlam, and Ellis (2009) had grown-up ESL learners to be exposed to a stream of input containing third person -s in grammar lessons while intended to show them an alternate grammatical feature.The intention was to investigate incidental achievement of third persons.All of the learners participating in the study were exposed to 51 illustrations of third persons which were the written data and 23 in oral form in two sessions enduring one hour in two days.Explicit and implicit knowledge tests were administered and demonstrated no procurement in either the short or long.
Three types of input-based instruction, namely (I) typographically enhanced input flood plus extensive listening and reading, (II) typographically enhanced input alone, and (III) input flood were compared by White (1998).This study found that all three types of input worked equally effectively in assisting French learners of English as their L2 to acquire the possessive pronouns 'his' and 'her', and it led the researcher to conclude that the target structure was equally perceptible in all three types of instruction.Williams and Evans (1998) analyzed the impacts of enhanced input i.e. comprising of a highlighted target structures in the text and also the same input in addition to instructing explicitly the learners and using the corrective feedback techniques for the learners to acquire English participial adjectives and present passive.In order to fulfill the purpose, a grammaticality judgment test and a sentence completion test were run.For the participial modifiers, the group which was exposed to explicit instruction and enhanced input outperformed both the enhanced input group and a control group on both tests while the distinction between the enhanced group and the control group was not statically significant.In the case of passive structure, both of the experimental groups demonstrated out performance over the control one on a sentence completion test; however no group contrasts was clear on the grammaticality judgment test.Subsequently, they reasoned that utilizing explicit instruction and input-based instruction at the same time might be more effective than input-based instruction alone.
Takimoto ( 2008) explored the impacts of unequivocal instruction and input-based instruction in light of the procurement of the semantic means for performing complex structures in which the explicit instruction took the manifestation of indirect awareness raising and the information based instruction comprised of organized data.
An alternate group was added to the experimental group that got immediate awareness rising and organized information.Each of the three groups performed better than the control group on a series of tests.A listening test was run as the deferred post-test.No significantly critical contrasts were identified between the test groups, with the exception of that the groups that got explicit instruction in addition to organized input performed less well than the other two groups on the test.Takimo recommended that this low execution may have been because in this condition the learners were presented to the unequivocal rules instead of needing to use them for themselves and the impact of this was not durable.It can be comprehended from this study that if the input based instruction makes the importance of the target structure clear, it is more viable than unequivocal instruction.
Through the wax and wane of implicit and explicit teaching, and the other balanced views of consciousness-raising in the aforementioned studies, one objective to follow in the present study is to investigate whether pre-presentation input flooding via reading has any significant effect on the acquisition of simple past tense.Another matter to find out is the possible effect of post-presentation input flooding via reading on the acquisition of past tense and the last is to see if there is any significant difference between the aforementioned techniques of input flood in terms of how they affect the acquisition process of past tense.
The present study made use of the flooded input through reading texts.The students were exposed to these materials before and after the explicit presentation of the target grammar.

Participants
Eighty male language learners from eight Race Three intact classes corresponding to elementary proficiency level and mostly Turkish as their L1 within the age range of 11-14 at Iran Language Institute (ILI) in Fall 2013 participated in the study.Race Three is the eighth level of the beginner levels at the ILI.According to the scale used at the institute, the learners who finish this level are assigned to register in elementary level if they are to go to the adults' department.Therefore, they were considered to have elementary proficiency and after the process of homogenization, 61 remained to be the main participants.

Instruments
To meet the objectives of the present study, a teacher-made pre-test and its parallel post-test on simple past tense with 30 items were administered.These tests enjoyed reliability indices of Alpha = 0.83 and Alpha = 0.72 respectively and were utilized as recognition devices.
This study also used dictogloss, a data collection tool, as a writing activity in which learners were asked to reconstruct the sentences about the pictures that were supposed to be explained by teachers first.This was applied in the study to see how much successful the learners are in producing the output in past tense other that recognizing it.
As for the treatment, the Reading Time sections of the learners' supplementary exercise books which contain numerous instances of simple past tense were adopted.

Procedure
To ensure the homogeneity of the participants, the learners' placement test scores and their average scores on three final test scores in the previous semesters were taken into account.Out of 80 participants, eight were found to be outliers and were excluded from the participants.Furthermore, in order to ascertain the participants' knowledge of the target grammatical structure, namely simple past tense, a pretest gauging simple past tense was administered.Therefore, the researcher included the participants who scored below 30 out of 50 to see if they have a good command in past tense or not.As a result, 61 participants remained to take part in the study.Following the balanced design in research, each group encompassed nearly equal N sizes for all the groups.Three homogeneous classes were randomly assigned into two experimental groups which had 21 and 20 learners, respectively; and one control group with 20 learners.
The intended treatment of pre-or post-presentation input flooding via reading was done in one session and dictogloss as a measurement tool was carried out in the following session to measure the achievement of the participants regarding simple past tense.This study was carried out in the first thirty minutes of twelve sessions for about two months.
Following the treatment, the researchers measured the target grammatical feature, past tense, through the dictogloss and immediate posttest.It is worth mentioning that the dictogloss was done under the pressure of 10-minute-time period to limit the thinking time and lessen the artificiality of the writing.
In G1, the teacher used a reading passage with the frequent past tense coverage prior to the presentation of the target grammar to determine whether any noticing happens to facilitate the learning of past tense.Also, to ensure the unity among the teachers and also to avoid the deviation from the ILI, the presentation of the grammar in the groups was a combination of deductive and inductive approach which was dominant in the ILI methodology.In G2, the teacher provided the learners with reading, including salient use of past tense, posterior to the presentation of the grammar just as a kind of practice rather than a noticing technique.Finally G3 as the control group of this study received none of the above mentioned techniques and followed just the normal process of presenting the grammar as stipulated by the respective language school.One-way between-groups analysis of covariance (ANCOVA) was conducted to compare the effectiveness of two different interventions designed to increase learners' past tense acquisition.The independent variable was the type of intervention (pre and post-presentation input flooding via reading) and control group, and the dependent variable consisted of learners' scores on test administered after the intervention was completed.Learners' scores on the pre-test were used as the covariate in this analysis.

Results
Before running ANCOVA, homogeneity of regression slopes test was used to check interaction between the covariate and the experimental manipulation.The result of this test is shown in Table 1.  4, the estimated mean differences of pre-presentation and control groups (10.63) is significant at (0.001) alpha level.Therefore, it may be concluded that pre-presentation input flooding via reading has positive and significant effect on the acquisition of the past tense.The estimated mean difference of post-presentation and control groups (9.74) is significant at (0.001) alpha level.Thus, the post-presentation input flooding via reading has positive and significant effect on the acquisition of past tense.Also, the results of table 4 reveal that the estimated means difference of pre and post-presentation groups (0.89) is not significant.According to this finding, it may be concluded that there is not any significant difference between pre-presentation inputs flooding via reading vs. post-presentation input flooding via reading in terms of their effect on the acquisition of the past tense.
One-way analysis of variance (ANOVA) was also conducted to compare the effectiveness of two different interventions designed to decrease learners' grammatical errors.In Table 5 groups' descriptive statistics have been reported.According to Table 5, the mean of grammatical errors of control group is higher than that of the post-presentation group.Therefore, post-presentation input flooding via reading has positive and significant effect on the achievement and production of past tense.
In this study, Levine's statistic to test homogeneity of variances wasn't significant (F = 0.05, p < 0.95).
According to this finding, it may be stated that the variance of grammatical errors is equal across groups.So the results of ANOVA are reported in Table 6.The results of Table 6 reveal that the F statistic of groups' differences is significant at 0.001 alpha level.The results of multiple comparisons (LSD post Hoc test) are reported in Table 7.According to Table 7, the mean difference of pre-presentation and control groups (2.57) is significant at (0.001) alpha level.The results of Table 5 show that the mean of grammatical errors of control group is higher than pre-presentation group.So according to this result, it may be concluded that the pre-presentation input flooding via reading group performs better than the control group and that out-performance is significant.The mean difference of post-presentation and control groups (1.96) is significant at (0.001) alpha level.It can also be found in table 7 that the estimated mean difference of pre and post-presentation groups (0.61) is significant at 0.006 alpha level.The results of table 5 show that the mean of grammatical errors of post-presentation group is higher than pre-presentation group.According to this finding, it may be concluded that the effect of the post-presentation input flooding via reading is higher than that of pre-presentation input flooding via reading on reduction of learners' grammatical errors.

Discussion
The first objective was to see if pre-presentation input flooding via reading had any significant impact on the acquiring of simple past tense.Therefore, first, the F statistics of group membership was calculated by means of ANCOVA to see if there was any significant difference among groups or not out of which the result was positive.Then, pair-wise comparison of groups' estimated mean scores indicated that pre-presentation input flooding via reading had a significant and positive impact on acquiring the simple past tense.This fact enhances the importance of noticing and consciousness raising activities in all classroom situations and especially in teaching grammar.This fact can also lead to the exercises which can provide some gaps and the students fill that gap.Noticing the gap may represent a higher level cognitive activity than the noticing of L2 examples, in that, in the former, the learner is making a cognitive comparison between a non-target-like form and a target like alternative and locating the gap between the two, whereas in the latter, no such comparison is being made (Saxton, 1997).
A dictogloss task was run in order to provide a data triangulation and also to determine whether the students gained some capabilities in the production of past tense.Therefore, F statistic of groups' differences through ANOVA was calculated first and proved that there is a significant difference between groups.Pair-wise comparison of groups' estimated mean scores proved that the difference is positive; i.e. input flooding via reading has a positive and significant effect on the acquisition of past tense.
The second dimension of the study was to find out if post-presentation input flooding had any significant effect on the acquisition of the past tense.Similar to the first objective, first, the F statistics of group membership was calculated through ANCOVA and the results proved that the intended treatment had a significant impact on the acquisition of simple past tense.Then, pair-wise comparison of groups' estimated mean scores showed that post-presentation input flooding via reading had also a significant and positive effect on the acquisition of past tense.
Through ANOVA analysis in dictogloss, the same results were proved.That is to say, not only did the students show progress in recognition but also they improved in the production of simple past tense.
The findings regarding this aspect can be applied in language classrooms and especially in grammar teaching.
Teachers and material designers can utilize activities in the classroom situations which make use of input flooding as a kind of practice to promote the acquisition of the intended target grammar.
The third and the last aspect of this study is concerned with the possible difference of the effects of pre-versus post-presentation input flooding via reading on the acquisition of simple past tense.In order to fulfill the purpose, the analysis of the results of both groups in posttest proved that both pre-and post-presentation groups performed fairly similarly.That is to say, there was no significant difference between pre-versus post-presentation input flooding via reading in terms how they might affect the acquisition process of simple past tense.The findings revealed that teachers can use both methods in their classrooms while presenting the target grammar.
However, analyzing the results of dictogloss demonstrated a different outcome.Through the one way ANOVA analysis, it was shown that pre-presentation group outperforms the post-presentation one.It confirms the fact that though the learners acted fairly similarly in the posttest.The ones exposed to pre-presentation input flooding produced more accurate output and their grammatical errors decreased during the course of instruction.Therefore, it can be claimed that the learners who are exposed to pre-presentation input flooding produced more accurate output than the ones in post-presentation group.
Similar to Trahey and White (1993) who reported temporary beneficial effect of input flood on SAV agreement with contrastive analysis French SVAO rule, this study in the first two questions confirmed the positive effect of input flood on past tense acquisition.However, due to lack of time, this study which used dictogloss corresponding to delayed posttest of Trahey (1996) may imply the permanent effect of input flood.The difference between these two studies can be attributed to the use of contextualized means like reading in input flooding.Therefore, through contextualized reading passages, this study, in contrast to Trahey andWhite (1993), andTrahey (1996), tried to tap a valid contextualized situation rather than to preoccupy the learners with just a pre-specified grammar point like SAV.
Contrary to Loewen, Erlam, and Ellis's (2009) study which dealt with incidental acquisition of third person -s and presenting a salient feature of indefinite article of generic reference, the present study operationalized the planned focus on form of past tense through timed written input flooding via reading.In contrast to Loewen et.al (2009) that used unequal number of written and oral flooded input and revealed no significant effect on the acquisition of third person -s, the current study made use of equal amount of input through aforementioned manipulation and proved its significant effect in the first two questions reinforced by the dictogloss results.
The comparison of White's (1998) three types of input-based instruction with the first two questions of this study showed an agreement in their effectiveness; however, the third question, disagreeing with the equal effect of White's treatments, determined the outperformance of pre-presentation input flooding group via reading over the other group.Hence, it can be inferred that contextualizing the input flood and its timing before or after presentation add to its noticing and its superiority over the existing input flooding and input enhancement techniques in the literature.
Similar to Williams and Evans (1998), this study used two types of measurement tools.The artificially increased highlighted target forms in one group of Williams and Evans' study corresponded to the post-presentation input flood via reading in this study.It is worth noting that they, like the present study, accompanied the input flood with instruction but without considering the timing factor which was included in this study and it distinguished the effect of pre-versus post-presentation; the fact evident in question three as shown in dictogloss task.Accordingly, the researcher may suggest the existence of a kind of learners' comparison of the target language with their interlanguage leading through the noticing and completing with instruction.

Conclusion
The researcher teachers concatenated the current theoretical issues of consciousness-raising, noticing, input enhancement and input flooding to operationalize the noticing of past tense with pre-presentation input flooding and practice effect with post-presentation input flooding.According to the findings, input flooding whether preor post-presentation makes significant difference and any of the two groups did not outperform the other.It can be inferred that since students come to the class with the presupposition of learning grammar until the level mentioned before (Race3), both groups performed nearly the same.Since the techniques utilized in this study were the innovative ones, they can act a triggering role in the novelty of the future research areas and be applied in the teacher-training-courses to add to its robustness.

Table 1 .
Result of homogeneity of regression slopesAs stated in Table3, F statistic of group membership is significant at 0.001 alpha levels, after controlling the pre-test scores (F = 155.31,p<0.001).This finding shows that there are significant differences among groups in post-test scores.The effect size statistic shows that group membership predicts 84 percent of variance in post-test scores.And according to table 3 in this study statistical power for group membership is 1, showing that sample size is sufficient for covariance analysis.Pair wise comparisons of groups predicting marginal mean scores are reported in Table4.

Table 4 .
Result of pair-wise comparison of groups estimated mean scores

Table 5 .
Descriptive statistics of groups' grammatical errors in dictogloss

Table 6 .
Result of variance analysis to compare groups in errors in dictogloss

Table 7 .
Result of pair-wise comparison of groups estimated mean scores