top of page
  • Writer's pictureSarah Alix

How Does Classical Conditioning Through use of Colour Signals Impact the Classroom?

Alexander Brown and Sarah Alix

How does classical conditioning through use of colour signals impact the classroom management of a secondary class? (A Trainee Project)


A key contributor towards a pupils’ classroom experience is how the teacher manages the class. For pupils to get the most out of their lessons they need to know that the teacher can manage the class and keep the lesson on track. There are a number of factors that contribute to the attitude of a class, including the subject, the individual teacher’s personality, the time of day, and the location of the classroom to name a few. Many of these cannot be changed. However, one factor that can be enhanced to promote a productive classroom atmosphere rooted in mutual respect is a consistently enforced set of rules (Payne, 2015). If pupils know the teacher’s expectations at any given time in the lesson, and understand the consequences associated with complying and not complying with them, the class can start feeling familiar and stable. While most schools have a standard reward and consequence system, each teacher is likely to differ in how readily they assign rewards and sanctions. In fact, there is likely to be a great deal of variety in how a single teacher uses the reward and consequence systems between lessons depending on their mood. While the same system is being used, the pupils do not always know what they must do to achieve a reward, or how much flexibility they have before their behaviour is deemed unacceptable and worthy of punishment. This inconsistency can cause pupils to kick back if they feel they have been treated unjustly or singled out; it also means compliant, well-behaved pupils can be overlooked when they should be rewarded. From a holistic perspective, it means the entire lesson is delayed as the teacher highlights why each instance of sanctioning does not meet expectations, or why each reward has been specifically given. This project investigated the effectiveness of a standardised method of classroom management, whereby different colours were visually displayed for different purposes: red when we expected the pupils to be silent and attentive, and green when we wanted pupils to be active and contributing to the lesson.

Literature Review

One of the most important and most emotionally-exhausting aspects of teaching is classroom management. Without the cooperation of the class it is impossible to teach a good lesson. Managing a class of individuals with different motivators and personalities is not a simple task; hence, individuals trained in specific means of managing behaviour substantially improve classroom management skills and the well-being of the educator (Dicke, Elling, Schmeck, & Leutner, 2015). This relationship between greater classroom management skills and reduced stress levels provided the motivation behind this action research project. There is then the question of which behaviour management strategies have the greatest effect. A study by Pas, Cash, O’Brennan, Debnam, & Bradshaw (2015) found that the most compliant classes were those where pupils have more opportunities to respond to the teacher and were subjected to less disapproval and negatively reactive management strategies. Thus, the intervention used in this research project was implemented with a focus on reducing the need for disapproving tones and encourage greater interactivity in lessons. Pas et al. (2015) also found that compliant classes typically had a greater proportion of white pupils. While this study did not specify whether the non-white pupils were also EAL (English as an additional language) pupils, this focus on class composition was considered when devising the intervention used in this research project. Further to EAL pupils, classroom management strategies should also take any pupils with special educational needs or disabilities (SEND) into account. In cases where individuals with autism spectrum Condition (ASC) are being introduced to new or desired behaviours visual cues are often used in tandem with instructions, either verbal or written (Ganz, Kaylor, Bourgeois, & Hadden, 2008; Ganz & Flores, 2010). However, Trembath, Vivanti, Iacono, & Dissanayake (2015) found no evidence that children with ASC are prominent visual learners and in fact found evidence contradicting this notion. An interesting result of this study was that typically developing children and children with global developmental delay did benefit from verbal instructions being supplemented with visual cues; only children with ASC did not improve under this intervention. A visual intervention was used in this research project to further investigate whether pupils with ASC reacted in a notably different way to the rest of the class.

Numerous recent studies have been carried out that highlight the existence of associative learning across a number of species. Hymenopterans, such as ants, bees, and wasps, appear to form particularly strong associations with between visual cues and extrinsic rewards (Bitterman, Menzel, Fietz, & Schäfer, 1983; Mancini, Giurfa, Sandoz, & Avarguès-Weber, 2018; Fernandes, Buckley, & Niven, 2018; Liefting, Hoedjes, Lann, Smid, & Ellers, 2018). Furthermore, early behaviourists such as Burrhus Skinner and Ivan Pavlov developed their theories after training animals, namely small mammals, through associative learning (MacBlain, 2018). If conditioning can be used to such great effect with animals that lack comprehension of language and verbal instructions, one would expect it would be of even greater use with regards to ‘training’ people. Conditioning humans is not a new concept, as MacBlain (2018) demonstrates. Operant conditioning, whereby positive and negative reinforcement measures are used to encourage desirable behaviour and discourage undesirable behaviour, has been ingrained into school behaviour for learning policies since the cane was common practice. Modern classroom management strategies rely heavily on operant conditioning in the form of extrinsic rewards and sanctions as positive and negative reinforcement respectively (Escobar & Nachev, 2017; Kelly & Pohl, 2018). While positive and negative reinforcement is utilised frequently in the classroom, very few of these modern behaviourist approaches use a memorable sensory signal to trigger a learned behavioural response. This research project combined the associative learning of Pavlov’s dogs with the reinforcement of Skinner’s operant conditioning to investigate whether subtle sensory signals, paired with positive and negative reinforcement, can be used to dictate the behaviour of students during a lesson.


According to Mertler (2016) action research is a “systematic inquiry” undertaken by the educator for the educator, with a focus on investigating aspects of one’s own teaching practice, how a school operates, or how students best learn. An educator may conduct research in their own classroom with their own students, consequently allowing them to implement a plan of action specifically designed with their own practice or pupils in mind. Despite this characterisation of action research, it is widely accepted that the term was introduced by Kurt Lewin as a means of “translating psychological principles and findings into practical recommendations for resolving social problems” following World War II (Lewin, 1946, cited in Stokols, 2006), and not initially applied to pedagogy. Therefore, a more accurate and concise definition of action research, given by Coghlan & Shani (2005), is “undertaking action and studying that action as it takes place”. In certain fields such as pedagogy and education, using an action research approach makes for a collaborative, productive, and useful investigation with applicable findings. An action research approach also encourages the researcher to critique themselves and reflect on the findings of the research in the interest of improving their practice. An extensive study undertaken by Davis, Clayton, & Broome (2018) found that action research projects provided student-teachers with a useful set of skills and had a positive impact on their teaching practice in the future, despite being deemed stressful and time-consuming at times.

We decided to apply this new intervention to a year 7 class as a part of ordinary teaching practice. This was because we felt we had a good relationship with the class, meaning they were unlikely to reject new rules out of defiance, but their tendency to talk through lessons and classroom management had been highlighted in previous observations. Furthermore, they were a class of 29, meaning we could test the efficacy of this intervention with a full class rather than a small group that may adopt new interventions more readily. The pupils were also relatively new to the school, and thus would have minimal prior biases towards the application of the reward and consequence systems. We checked beforehand and ensured that none of the pupils suffered from colour deficiency that may make distinguishing particular colours difficult. This project was accepted with no revisions by a University ethics board.

We decided to use the colours red and green to represent a desire for silent attention and active participation respectively, as these colours have similar “stop” and “go” connotations in other aspects of day-to-day life, such as traffic lights. We made sure to always use the left hand to display the red signal and the right hand to display the green signal; consequences and rewards were written on the corresponding sides of the whiteboard. This was to account for any colour deficiency or colour blindness that we had not been made aware of, and further promote structure and routine for the benefit of the ASC pupils in the class. When using a red signal, we always used a countdown or a call for attention alongside it, and initially held it aloft. When a pupil was given a sanction for talking while the red signal was on display, we would always reiterate that we should not hear any voices other than those of pupils we have chosen to answer. When using a green signal, we responded to all answers with verbal praise and bounced questions to other pupils far more frequently so to promote a classroom discussion. Where another member of staff addressed the class and sought a response, we also displayed the green signal. These actions helped the pupils build an association between the displays and the corresponding consequences, but at no point did we explicitly tell the pupils what each signal meant.

We used the signals for five weeks in total. Each week consisted of two lessons. Each week we reduced the strength of the signals. This was working on the logic that, as pupils begin to build an association between the signal and the desired behaviour, the need for a conspicuous should lessen until a subtle signal achieves the desired response. On the first week we used A4 sheets of paper, one red and one green, as my signals. In this way we ensured that all pupils could register the presence of this new signal, even though they were given no context behind it. With each consecutive week we reduced the conspicuousness of the signal; on the second week we folded the sheets to A5 size, on the third week we folded the sheets to A6 size, on the fourth week we wrapped two whiteboard pens in the coloured papers, and on the fifth week we simply used red and green dry-wipe pens as the signals. The expectation was that, as the pupils became more used to the coloured signals, the strength of the signal could be reduced and still achieve the desired effect.


The quantitative data of sanction and achievement point counts as our primary data, with qualitative observations and reflections as secondary data to support the findings. The trends displayed by the quantitative data have then been cross-referenced with the qualitative data to highlight any consistent patterns or inconsistencies.

From the baseline average number of sanctions, taken from lessons prior to the introduction of the signals, the number of sanctions per lesson fluctuates for the first three weeks. This is reflected in the observation form which indicates that the pupils were still adjusting to the signals for this period. The number of rewards also showed an initial decrease in the first week, again suggesting a lag in the uptake of these signals, but seemed to stabilise by the second or third week. After the second week the number of rewards per lesson gradually tended towards the initial baseline. From the ratios of negative to positive comments on both the self-reflections and observation forms, it is evident that classroom management seemed weaker and less consistent on the third week than the second and fourth weeks. This is consistent with the numerical data, corresponding with the peak in sanctions given out between the second and third weeks that disturbs an otherwise gradually decreasing trend.

Sanctions reached their lowest mean frequency on the fourth week and remained at that level for the fifth week. The ratios of negative and positive comments from self-reflections and observation forms for these weeks support this trend, with both ratios reaching their most positive on the fourth week and the ratio for self-reflection comments remaining this way throughout the fifth week. Rewards (AP) also appeared to decrease during these final weeks.

By the end of the research project the average number of sanctions per lesson had decreased by approximately two. This meets one of the expected outcomes of this research project. Contrastingly, the average number of achievement points given out per lesson showed a sharp peak in the second week but had almost returned to the prior average frequency by the end of the project. This did not meet the expected outcomes of this project and indicates differences in the integration of the red and green signals.


The predicted outcomes of this research project were that there would be a sharp increase in the number of sanctions and a sharp decrease in the number of rewards on the first week that the signals were introduced. This was expected because the pupils had no prior exposure to the colour signals and, without any explicit explanation of what each signal meant, they were bound to infringe upon the rules and incur some negative reinforcement. The observed response during week one aligned with these predictions; a greater number of sanctions and lower number of rewards were given out during the first week of using the colour signals. Observations and reflections indicate that the colour signals were being used to effectively manage behaviour, which is an improvement from comments from prior weeks, evidence showed that, after being exposed to the colour signals for only a single lesson, the pupils had started to modify their behaviour for the next lesson.

This is consistent with the tone presented in the self-reflection for week four, where we have stated that I felt “that behaviour was largely positive,” despite a reduced frequency of rewards. From the comments on the observation form for week four, such as “[good] use of rewards,” it is evident that positive reinforcement was still being used and recognised. However, this subconscious increase in expectations and decrease in positive reinforcement should be considered with caution. Articles such as those by Corr (2002) and Fisher, Thompson, Hagoplan, Bowman, & Krug (2000) explain that delayed rewards, or rewards that do not measure up to magnitude expected, can result in frustration and aberrant behaviour. Future applications of this or similar interventions should take care to persist with positive reinforcement to maintain associations between the signals and rewards for compliant behaviour.

Only two pupils consistently accrued sanctions regardless of the intensity of the signal. Interestingly, one of these pupils is diagnosed both ASC and ADHD, and the other exhibits some similar behaviour which we have previously noted as ADHD tendencies. Trembath et al. (2015) that suggested that pupils with ASC do not form visual associations as readily as other individuals. It also provides evidence that supports the findings of Furukawa et al. (2018) on the behavioural sensitivity of children with ADHD to changes in reward availability. Since both signals entailed different consequences for some of the same behaviours, namely calling out or answering aloud, it is our belief that these struggled to adapt to these different signals.

There were some limitations to the study that should be considered and that we will be aiming to overcome in future applications of this intervention. Firstly, we found we were prone to raising the red signal and calling for attention when we wanted silent attention but did not always display the green signal if they were permitted or even encouraged to actively participate in classroom discussion or the current activity. This meant that the red signal was recognised far more readily, and its meaning understood far better, than the green signal. This therefore meant that the number of sanctions given out represented the efficacy of these signals far more than the number of rewards. Furthermore, it limited how much positive association there was with following the signals and relied heavily on the negative association with disobeying them. Secondly, the results regarding the efficacy of this intervention are confounded somewhat by previous actions used to call for attention, namely raising my hand and waiting for them to silently mimic the action. This meant that the pupils may have already been inclined to fall silent and pay attention when we raised my hand, whether we were holding a red signal or not. This became less of a limitation later on when we did not hold the signals aloft as much but is possibly responsible for the rate at which pupils started reacting to the red signal. Lastly, as the quantitative data shows, despite using these signals to regulate my use of sanctions and rewards, my use of rewards and positive reinforcement decreased as the need for sanctions decreased. While subconscious, this is evidence that there is still a bias that leads teachers to give out more rewards in a class with a lot of disruptive behaviour to highlight desirable behaviour than in a class where all pupils are well behaved and meeting expectations. To effectively implement an intervention such as this, the green card needs to be used and acted upon as readily as the red card so that pupils are inclined to make use of the opportunity to achieve rewards, rather than just trying to avoid sanctions.


Overall the evidence shows that after five weeks under this conditioning the amount of non-compliant behaviour, and hence the need for negative reinforcement through sanctions, was reduced. The qualitative data also indicates that compliant, positive behaviour increased, but the quantitative data shows a gradual decline in the number of rewards given, suggesting a possible increase in expectations. Thus, the colour signal intervention had a positive effect of pupil behaviour and classroom management and can seemingly be employed alongside the school’s sanction and reward policy to remind pupils of the teacher’s expectations and condition their behavioural response. Changes will be made to this intervention when it is next put into action. The greatest change we intend to make is to explain what the cards mean to the pupils before we use them. The reason behind this is that, whatever the reason, the pupils with ASC and ADHD did not respond to the signals as the rest of the class did. We are therefore curious as to whether this was as a result of me constantly changing cards, and therefore shifting the parameters of what is acceptable and what is not at any given time, or whether they simply did not make the associations between the visual signals and the desired behaviours. By telling the next class what each card means we can develop a better understanding of which of the above explanations is most likely from the response of any ASC pupils in the new class. We also felt that the strength of the signal may have decreased too rapidly. Thus, we will write sanctions in red pen and rewards in green pen on the board. This will hopefully strengthen the association between the different colours and their associated consequence. Lastly, a confounding factor that could not be avoided in this research project was the existence of prior classroom management strategies and routines. Hence, we intend to introduce this intervention to the next class after only two weeks of being taught. This limits how many other routines the pupils can become accustomed to before they are introduced to the colour signals.


Bitterman, M. E.; Menzel, R.; Fietz, A.; and Schäfer, S. (1983) ‘Classical conditioning of proboscis extension in honeybees (Apis mellifera)’, Journal of Comparative Psychology, 97(2), pp. 107-119.

Coghlan, D. and Shani, A. B. (2005) ‘Roles, politics, and ethics in action research design’, Systemic Practice and Action Research, 18(6), pp. 533-546.

Corr, P. J. (2002) ‘J. A. Gray’s reinforcement sensitivity theory and frustrative nonreward: a theoretical note on expectancies in reactions to rewarding stimuli’, Personality and Individual Differences, 32(7), pp. 1247-1253.

Davis, J.; Clayton, C.; and Broome, J. (2018) ‘Thinking like researchers: action research and its impact on novice teachers’ thinking’, Educational Action Research, 26(1), pp. 59-74.

Dicke, T.; Elling, J.; Schmeck, A.; and Leutner, D. (2015) ‘Reducing reality shock: The effects of classroom management skills training on beginning teachers’, Teaching and Teacher Education, 48, pp. 1-12.

Escobar, M. C. and Nachev, S. (2017) ‘A study of the way five teachers make decisions in the “EFL Classroom”’, Colloquia, 4, pp. 157-187.

Fernandes, A. S. D.; Buckley, C. L.; and Niven, J. E. (2018) ‘Visual associative learning in wood ants’, Journal of Experimental Biology, 221(3), pp. 1-8.

Fisher, W. W.; Thompson, R. H.; Hagopian, L. P.; Bowman, L. G.; and Krug, A. (2000) ‘Facilitating Tolerance of Delayed Reinforcement During Functional Communication Training’, Behavior Modification, 24(1), pp. 3-29.

Furukawa, E.; Alsop, B.; Caparelli-Dáquer, E. M.; Casella, E. B.; da Costa, R. Q. M.; Queiroz, P. M.; Galvão, P. A.; Benevides, L. R. D. S.; Jucá-Vasconcelos, H. P.; and Tripp, G. (2018) ‘Behavioral adjustment to asymmetric reward availability among children with and without ADHD: effects of past and current reinforcement contingencies’, ADHD Attention Deficit and Hyperactivity Disorders, pp. 1-10.

Ganz, J. B. and Flores, M. M. (2010) ‘Implementing visual cues for young children with autism spectrum disorders and their classmates’, Young Children, 65(3), pp. 78-83.

Ganz, J. B.; Kaylor, M.; Bourgeois, B.; and Hadden, K. (2008) ‘The impact on social scripts and visual cues on verbal communication in three children with autism spectrum disorders’, Focus on Autism and Other Developmental Disabilities, 23(2), pp. 79-94.

Kelly, J. and Pohl, B. (2018) ‘Using Structured Positive and Negative Reinforcement to Change Student Behavior in Educational Settings in Order to Achieve Student Academic Success’, Multidisciplinary Journal for Education, Social and Technological Sciences, 5(1), pp. 17-29.

Liefting, M.; Hoedjes, K. M.; Lann, C. L.; Smid, H. M.; and Ellers, J. (2018) ‘Selection for associative learning of color stimuli reveals correlated evolution of this learning ability across multiple stimuli and rewards’, Evolution, 72(7), pp. 1449-1459.

MacBlain, S. (2018) Learning Theories for Early Years Practice. London: SAGE Publications.

Mancini, N.; Giurfa, M; Sandoz, J. C.; and Avarguès-Weber, A. (2018) ‘Aminergic neuromodulation of associative visual learning in harnessed honey bees’, Neurobiology of Learning and Memory, 155, pp. 556-567.

Mertler, C. A. (2016) Action Research: Improving Schools and Empowering Educators. 5th edn. California: SAGE Publications.

Pas, E. T.; Cash, A. H.; O’Brennan, L.; Debnam, K. J.; and Bradshaw, C. P. (2015) ‘Profiles of classroom behavior in high schools: Associations with teacher behavior management strategies and classroom composition’, Journal of School Psychology, 53(2), pp. 137-148.

Payne, R. (2015) ‘Using rewards and sanctions in the classroom: pupils’ perceptions of their own responses to current behaviour management strategies’, Educational Review, 67(4), pp. 483-504.

Stokols, D. (2006) ‘Toward a science of transdisciplinary action research’, American Journal of Community Psychology, 38(1-2), pp. 63-77.

Trembath, D.; Vivanti, G.; Iacono, T.; and Dissanayake, C. (2015) ‘Accurate or assumed: Visual learning in children with ASD’, Journal of Autism and Developmental Disorders, 45(10), pp. 3276-3287.

61 views0 comments

Recent Posts

See All

Pathological Demand Avoidance (PDA)

Pathological Demand Avoidance (PDA) What is Pathological Demand Avoidance? Pathological Demand Avoidance (PDA) is the avoidance of some of the everyday demands in life. For example, avoiding homework,


bottom of page