Validating fNIRS through a brand perception experiment: a promising approach

Validating fNIRS through a brand

perception experiment: a promising

approach

Bachelor Thesis Psychobiology

Name: Dianne M. Urbach

Student number: 11671114

Supervisor: A.R. Van der Leij

Second corrector: M. Hashemi

Date: 3-7-2020

Abstract

Brand perception appears to be relevant to humans, as they value the image they perceive of a brand. Consumer neuroscience is a branch of neuroscience that studies consumer behavior by means of brain research. Much has been devoted to study people’s opinions on brands. Previous research has found that the neural basis of brand preference is the dorsolateral and ventromedial prefrontal cortex. Within consumers research an emerging neuroimaging technique is functional near infrared spectroscopy (fNIRS), a portable and low-cost approach to measure brain activity. Resulting from previous research, the question emerges, to what extent is fNIRS an adequate approach to measure the preference of brand logos, which consumers define to be positive in comparison to brands which consumers define to be negative. Arising from previous research, it is hypothesised that fNIRS is a qualified approach to measure brain activity and that it enables the researcher to perceive the brand preference of participants. In order to test the hypothesis, thirty subjects were recruited to participate in a brand perception experiment. The participants were asked to observe positive or negative self-defined brand logos while their neural activity was measured utilizing fNIRS. The results present significant differences between perceiving a positive and negative defined brand logo in both the vmPFC and dlPFC. These findings suggest that fNIRS indeed is a promising approach to measure neural activity and that it can successfully distinguish the perception of positive and negative defined brand logos.

Key words: functional Near Infrared Spectroscopy, brand perception, brand preference, ventromedial prefrontal cortex, dorsolateral prefrontal cortex, brain–computer interface

Introduction

The perception of brand logos has been a subject of discussion for ages, but the question remains: do human-beings actually care about their perception on a brand? A memorable moment in history, called ‘The Cola Wars’ suggests we do so (The Coca-Cola company, 2020). The Coca-Cola company wanted to improve their image, so they introduced an improvement of the old recipe, and branded it ‘new Coke’ in 1985. Instead of people reacting positively to the sugared drink and the new brand logo, they experienced the biggest backlash known in the history of marketing. People were outraged by the change of the brand and wanted back ‘the Real Thing’. They argued that the Coca-Cola brand was not the same anymore. Due to this, the Coca-Coca-Cola company went back to their old recipe and logo. Because of people’s seemingly conditioned opinions on the brand’s image, brands urge to determine whether their business concept is pleasing to customers. This shows that brand image is of value to people’s opinions other than the product on its own. Brand perception thus appears to be relevant to humans.

In previous research, it has been investigated whether rhesus macaques show choice behaviour that is similar to humans in response to sex and social status in advertising. In this study by Acikalin, Watson, Fitzsimons and Platt (2018), monkeys were shown brand logos either in combination with a sexual picture - an image of the genitals of a monkey from the opposite sex - or in combination with a neutral picture. When the monkeys were made to choose between two logos, the monkeys appeared to choose the logo that was previously shown in combination with the sexual image over the logo shown in combination with the neutral image. Furthermore, no reward was involved during the time of choosing the logo, thus no material association was provoked. This demonstrates that monkeys are able to be conditioned to develop a preference for a brand logo. A brand logo is for them a meaningless image, until a positive association is formed between the logo and a sexual image. It raises the question whether humans are likewise conditioned to develop a preference for brands, for instance through advertising.

Consumer neuroscience is a branch of neuroscience that studies consumer behavior by means of brain research. Much has been devoted to study people’s opinions on brands. In the review of Plassmann, Ramsøy and Milosavljevic (2012) multiple studies are cited that investigated whether humans are conditioned to prefer certain brands and which brain areas contribute to this brand preference. They concluded that humans, resembling monkeys, are conditioned for brand preference through advertising. Moreover, they concluded that the ventromedial prefrontal cortex (vmPFC) and the dorsolateral prefrontal cortex (dlPFC) are the neural basis for brand preference. For example, in the study of Koenigs and Tranel (2008), healthy participants and participants with vmPFC lesions were asked to perform a blind and semi-blind taste-test concerning cola. The participants seemed to prefer Pepsi during a blind taste test, whereas they seem to prefer Coca-Cola during a semi-blind taste-test where they perceive the brand of the cola they are drinking. This phenomena is called the ‘Pepsi paradox’. However, the patients with vmPFC lesions disclosed no Pepsi paradox. The vmPFC lesion patients preferred Pepsi during the blind taste-test and also during the semi-blind taste-test. Thus, the researchers derived that the vmPFC is an essential neural substrate for brand perception. Additionally, a research considering decision making, from Martin and Lawrence (2003), found that the vmPFC is activated during preference judgement. This is in line with other studies suggesting that the vmPFC plays an essential role in brand preference (e.g. Schaefer, Berens, Heinze & Rotte, 2006; Goel & Dolan, 2001).

In addition, other studies have investigated the effect of the dlPFC on brand preference in the human brain. For example in the study of Camus et al. (2009), the authors researched whether the dlPFC has a causal effect to decision making and valuation. The participants in this study were

asked to indicate the value they acknowledge to a food product in a monetary unit. They found that when they effectively killed the activity of the dlPFC using rTMS, the participants experienced a decrease in precepted value to the stimuli, in contrast to the control condition. This is in line with other investigations concerning the dlPFC and brand perception (e.g. Klucharev, Smidts & Fernández, 2008). These findings suggest that the prefrontal cortex, specifically the dlPFC and vmPFC, is the neural basis of brand preference.

Within consumers research, an emerging neuroimaging technique is functional Near Infrared Spectroscopy (fNIRS). The technique is a portable and low-cost approach to measure brain activity. fNIRS can contribute to brain research by broadening the communication methods of neural signals through a brain-computer interface (BCI). fNIRS is a technique that detects changes in the amount of oxygenated (O2Hb) and deoxygenated (HHb) haemoglobin molecules in blood, named

haemodynamic response (Leon-Carrion & Leon-Dominguez, 2012). Earlier research indicates a link between haemodynamic response and brain activity, titled neurovasculair coupling (Arthurs & Boniface, 2003). Due to the resemblance of fNIRS to functional Magnetic Resonance Imaging (fMRI) it can be argued that fNIRS is a reliable, yet indirect, manner to measure brain activity. The near infrared light spectrum ranges from 700 to 900 nm. In this ‘optical window’ O2Hb and

HHb can be perceived conjointly, whereas water cannot be detected and thus cannot obstruct the measurements (Jöbsis, 1977). A fNIRS set-up contains near infrared light sources (LEDs) and light detectors. A light source emits near infrared light through the scalp, into the cortex and is received by a light detector. The emitted light composes a banana shaped path to the photo-detector (figure 1) (Gratton, Maier, Fabiani, Mantulin & Gratton, 1994).

Figure 1. Schematic apprehension of the banana shaped deflection of near infrared light (Leon-Carrion & Leon-Dominguez, 2012).

fNIRS has many advantages in comparison to other brain imaging methods, such as fMRI. fNIRS is non-invasive, portable and inexpensive (ICNIRP, 2006). Inexpensive brain imaging methods have the first advantage to be utilized in all types of research, especially those receiving less funds. Secondly, fNIRS is portable, making it easier to conduct brain research on patients who have trouble sitting or are bedridden (Irani, Platek, Bunce, Ruocco & Chute, 2007). Furthermore, fNIRS has the ability to expand neuroscience outside of the western world, this may give a more realistic and inclusive view of neuroscientific evidence (Burns et al., 2019). Additionally, during fNIRS monitoring, patients are able to move. This could be beneficial for research during, for example, an epileptic seizure or other medical disorders which causes uncontrollable movement. Thirdly, fNIRS

has a high temporal resolution in the order of milliseconds, exceeding the temporal resolution of fMRI (Huppert, Hoge, Diamond, Franceschini & Boas, 2006). While fNIRS has many advantages, some disadvantages need to be taken into account. For example, fNIRS has a reduced spatial resolution in comparison to fMRI (Irani et al., 2007). In addition, another weak component of fNIRS, described in the same research, is the inability to measure the deep brain. Thus, fNIRS has advantages compared to other neuroimaging techniques, however fNIRS also faces some disadvantages that need to be taken into account.

fNIRS has been an article of interest in neuroscientific research. Hosseini and colleagues (2011) conducted an experiment, in which they examined whether personal preference for a visually presented object could be predicted, on the basis of fNIRS data. They found that they could correctly predict this, while monitoring the anterior frontal cortex using fNIRS. This result suggests that fNIRS is a suitable neuroimaging technique to assess imaginary preference in the human brain, however further research is necessary to confirm this statement.

Resulting from previous research presented in the introduction, the question emerges, to what extent is fNIRS an adequate approach to measure the preference of brand logos, which consumers define to be positive in comparison to brands which consumers define to be negative. Therefore, in this study, it will be investigated whether the preference of participants on brand logos can be assessed using fNIRS. Investigating whether fNIRS is an adequate approach to measure brain activity, can contribute to brain research in the field of neuroscience, including clinical and consumers research domains. Arising from the research of, among others, Plassmann and colleagues (2012) and Hosseini and colleagues (2011), it is hypothesised that fNIRS is a qualified approach to measure brain activity and is able to perceive brand preference of participants. In this study, measurements of the prefrontal cortex (PFC), acquired by functional Near Infrared Spectroscopy (fNIRS), will be utilized to derive the prefrontal activation in humans whilst observing a brand logo. To research this, participants will be visually presented with a brand logo, meanwhile fNIRS measurement will be conducted. After participating in this experiment, the participants completed a questionnaire in which they indicated which brand logos they define to be either positive or negative. In the data analysis section these results are evaluated. Proceeding from earlier research, as discussed above, it is expected that the activation in the vmPFC and dlPFC will increase whilst observing a brand logo, when participants pronounce the preference for that specific brand in the questionnaire.

Materials & methods

Subjects

A total of 32 Dutch-speaking, adult subjects (ages 19-60, mean = 23.91, SD = 8.74; 19 females) participated in this experiment. The participants received a payment, for their participation in this experiment. Two participants were excluded, due to fNIRS measurement failure. From each participant, a written informed consent was obtained previous to the onset of the experiment. The informed consent was written regarding the guidelines by the board of ethics from the psychology department of the University of Amsterdam.

Stimuli

For this experiment, 150 Dutch television commercials were selected by Neurensics, a neuromarketing company that conducts brain research for marketing purposes. The commercials were selected on grounds of won awards or computed brain response. Each television commercial pertained a brand logo. In the experimental trials, the displayed logos were identical to the logos

that appeared in the commercials (Appendix D). Additionally, 25 brand logos were displayed of which no commercial was adjoined, to function as a control. All brand logos were rendered to a 1400x788 pixels format.

Procedures

The experiment was manufactured in EventIDE (Okazolab Ltd, Delft, The Netherlands). The experiment is a brand perception task and consisted of three separate parts, the first and third part are visualized in figure 2.

Figure 2. In this figure a representation of the experimental procedure of the first and third block, is visualised. The first screen containing a brand logo is shown for two seconds, after which two words appear beneath the logo. Thereafter a fixation cross is shown for the remaining time of the trial. In a block this visualised task flow is repeated three times for all 25 logos.

In the first part of the experiment, the participants were shown a brand-logo for two seconds. After the two seconds elapsed, two Dutch words appeared underneath the brand-logo. The words that appeared beneath the brand logos were either positive (i.e. 'sympathiek', 'betrouwbaar'), neutral (i.e. 'herkenbaar', 'nietszeggend') or negative (i.e. 'afstotelijk', 'onoprecht') (Hermans, De Houwer & Eelen, 1994). The words appeared at random in the combinations negative, positive-neutral and negative-positive-neutral. At the time the words appeared, the participant was asked to indicate which of the two words they found best fitting to the brand logo that was shown. They could press either the keypress ‘Z’ for the left word or the keypress ‘M’ for the right word. The participants were instructed to press no key when they had no opinion on the brand logo. After a key was pressed or after a period of 2.5 seconds a blank screen followed. This blank screen was displayed for a period of time until the interstimulus interval (ISI) of approximately seven seconds (± one second) was reached (figure 2). During the first part, 29 brand-logos were shown three times, of which 25 brand logos were the experimental condition and four brand logos were the control condition. The four brand logos of the control condition were brand logos of which no commercial was shown in the second part of the experiment. These serve as a control to measure the standard change in neuronal activity, without the manipulation of the commercial. After every 29 trials, a self-imposed break was initiated.

The following part of the experiment consists of presented television commercials, with an average of 40 seconds longitude. The 25 television commercials, corresponding to the experimental brand logos, were displayed. A pause of eight seconds was introduced after every commercial and after every five commercials a self-imposed break was initiated.

The last part is equivalent to the first part (figure 2). The brand-logos were displayed anew, in a random order, and the participants had to decide which word fits better to the brand-logo that was presented.

fNIRS data acquisition

During the execution of the experiment, the participants were assembled with the Brite 24, which acquires data using Oxysoft software (Artinis; NIRx Medical Technologies, Elst, The Netherlands). The Brite 24 is a portable fNIRS measuring system, which acquires the O2Hb and HHb levels. The

fNIRS systems measures with wavelengths of 758 nm and 853 nm and the data was acquired with a sampling rate of 125 Hz. The fNIRS setup operated with ten near infrared light sources (Tx1-Tx10) and eight light receivers (Rx1-Rx8), located on the PFC (figure 3; Appendix A) (D'esposito, Cooney, Gazzaley, Gibbs & Postle, 2006). The optodes are placed within an 30 mm radius from a light transmitter to a light receiver. Consequently, neural activity is collected at a depth of approximately 15 mm (Ferrari, Mottola & Quaresima, 2004). No channel could be formed between Tx2-Rx1 (channel 4) and Tx7-Rx5 (channel 26), due to the distance between these optodes exceeding the 30 mm radius. For a specification of all channels see Appendix B.

The channels selected for the data acquisition of the vmPFC and the dlPFC, are presented in table 1. The locations of the dlPFC and the vmPFC are based on Brodmann’s Atlas (Brodmann, 1909; Brodmann, 1912). LEFT RIGHT VMPFC 36 (Rx7 - Tx10) 22 (Rx3 - Tx5) 44 (Rx8 - Tx10) 24 (Rx4 - Tx5) DLPFC 28 (Rx5 - Tx6) 2 (Rx1 - Tx1) 32 (Rx7 - Tx7) 8 (Rx2 - Tx1)

Table 1. Selected channels for the measurements of the vmPFC and the dlPFC, based on Brodmann’s Atlas (Brodmann, 1909; Brodmann, 1912).

Figure 3. Schematic representation of the fNIRS optodes placement. The light sources are indicated with a yellow colour (Tx1-Tx10). The light receivers are indicated with a blue colour (Rx1-Rx8). The channels are indicated with a green line.

Questionnaire

After participating in the experiment, the participants received an e-mail containing a questionnaire. In this questionnaire, the participants were asked to indicate whether they are interested in the brands they have seen. They could choose between the options: interested, not interested and no opinion. The same format was used for the questions: whether they would purchase products from the brands they had seen, whether they would recommend the brands they had seen and whether they reckon the brand they had seen contributed positively to the environment (Appendix C). The count of the answers from these questions resulted in the baseline assessment for the experiment.

Data processing

The data was analysed in Matlab (The Mathworks, Inc., R2018a). The raw fNIRS signals were smoothed by applying a 5th order cubic Savitzky-Golay filter with a frame length of 12 seconds. Next, the smoothed signal was filtered with a Chebyshev Type 1 lowpass filter (passband frequency 2 Hz, stopband frequency 5 Hz, Passband ripple 1dB, stopband attenuation 20dB). For analysis, a stick model timeseries was generated and convoluted with a canonical BOLD function. The resulting convolved model with 2 was regressed against the standardised and mean-centred timeseries using a glmfit, resulting in 2 beta's per channel.

Statistical data analysis

The further statistical analysis was performed in RStudio (R version 4.0.1) was used. The dependent variables, the positively and negatively precepted brand logos derived from the questionnaire, and the independent variable, the O2Hb concentration, were compared using a

paired t-test. Beforehand, to test for the assumptions of normality, a Shapiro-Wilk test was executed. When the assumptions for the paired t-test were not met, a non-parametrical Wilcoxon test was enacted.

Results

The organisation of the channel pattern localisation results are envisioned in figure 4A. The numbers in this table correspond to the number of each channel and are placed on the PFC as visualized in figure 4B. The results from the channel pattern localisation are shown in figure 4C for the positively defined brand logos, and in figure 4D for the negatively defined brand logos. A clear distinction in the amount of oxygenated haemoglobin is visible between a positively and negatively defined brand logo.

Figure 4. The purpose of this figure is to give an illustration of the localisation of O2Hb in the PFC, whilst

observing a positively and negatively perceived brand logo. A, B. A schematic representation of the channel pattern localisation, on the frontal regions of the participants. C. Pattern of the amount of O2Hb for positive

brand perception. D. Pattern of the amount of O2Hb for negative brand perception. An evidently larger amount

of O2Hb is visible for the positive condition as compared to the negative condition.

In figure 5, the medians of the results visualised in figure 4 are plotted. The results do not match the assumptions of normality and therefore a non-parametrical test was used to compare the conditions. The results show no significant difference in O2Hb concentration between the positive

and negative condition [V = 321; p-value = 0.069]. As follows, the medians of the oxygenated haemoglobin in the PFC do not differ between observing a positively perceived brand logo and observing a negatively perceived brand logo.

B

.

C.

.

D.

Prefrontal activity while

perceiving positive logos

Prefrontal activity while

perceiving negative logos

Figure 5. This figure illustrates the results of the brand perception data. Displayed are the medians of the O2Hb in the PFC whilst observing positively and negatively precepted brand logos. The medians show no

significant difference (V = 321; p-value = 0.069).

The means of the oxygenated haemoglobin in the vmPFC, while perceiving a positive brand logo and a negative brand logo, are visualised in figure 6A. The assumptions for a parametrical t-test were met in both samples. The results show a significant difference between the means [t = 2.086; df = 29; p-value = 0.046]. The means of the O2Hb concentration in the dlPFC, when observing a

positive and a negative brand logo, are presented in figure 6B. The results indicate a significant difference between the means [t = 2.052; df = 29; p-value = 0.049]. Hence, for both the vmPFC and the dlPFC, the amount of oxygenated haemoglobin results to be higher whilst observing a positively perceived brand logo in comparison to a negatively perceived brand logo.

Figure 6. This figure illustrates the results of the brand perception data specifically for the vmPFC and the dlPFC. A. Displayed are the means of the O2Hb whilst observing positively and negatively defined brand logos

in the vmPFC. The means display a significant difference (t = 2.086; df = 29; p-value = 0.046). B. Displayed are the means of the O2Hb whilst observing positively and negatively defined brand logos in the dlPFC. The

means in this figure appear to differ significantly (t = 2.052; df = 29; p-value = 0.049).

Discussion

The aim of this study was to evaluate to what extent fNIRS is an adequate approach to measure the preference of brand logos in humans. To research this, participants were asked to look at brand logos while their neural activity was recorded using fNIRS. In this study, specifically the neural activity in the dorsolateral prefrontal cortex and the ventromedial prefrontal cortex were observed. The results obtained in this study show that the activity in the dorsolateral prefrontal cortex and ventromedial prefrontal cortex increases significantly whilst observing a positively defined brand logo in comparison to a negatively defined brand logo. Whereas, the overall activity of the prefrontal cortex does not show this significant difference between conditions. These results indicate that fNIRS is a promising approach to measure neural activity and can successfully distinguish the perception of positively and negatively defined brand logos. In general, from these results fNIRS appears to be an adequate technique to measure neural activity.

The current results are in line with previous research on brand perception. Firstly, both the ventromedial prefrontal cortex and dorsolateral prefrontal cortex appear to differ significantly in oxygenated haemoglobin levels between observing positively and negatively defined brand logos. This finding is in line with previous research, which concluded the ventromedial prefrontal cortex and dorsolateral prefrontal cortex are the neural basis of brand perception (Plassmann et al., 2012).

Secondly, the comparison between the oxygenated haemoglobin levels in the prefrontal cortex, whilst observing positively and negatively defined brand logos, did not appear to differ significantly. This finding might be due to the fact that the prefrontal cortex is not wholly involved whilst observing a brand logo. Evidence has been obtained in previous research for the claim that the prefrontal cortex contains multiple regions with distinctive functions (Gilbert, Henson & Simons, 2010). As is visible in figure 4, regions within the prefrontal cortex occur to differentiate from each other, instead of revealing corresponding activity.

Thirdly, as discussed earlier, the main finding of the research from Martin and Lawrence (2003) is that the ventromedial prefrontal cortex is the neural basis of preference judgement. Preference judgement is an important component in decision making. Other studies concluded the same on the ventromedial prefrontal cortex (e.g. Goel & Dolan, 2001; Schaefer et al., 2006; Koenigs & Tranel, 2008; Plassmann et al., 2012). This might explain why the results obtained in this study show a greater difference between the positive and negative conditions when observing the ventromedial prefrontal cortex in comparison to the dorsolateral prefrontal cortex. A smaller difference between the oxygenated haemoglobin for the conditions became apparent for the dorsolateral prefrontal cortex. Fewer evidence is available to support the claim that the dorsolateral prefrontal cortex is an essential area for brand preference from previous research. While there is existing evidence for the involvement of the dorsolateral prefrontal cortex in brand preference, there is significantly less evidence in comparison to the ventromedial prefrontal cortex (e.g. Klucharev et al., 2008; Camus et al., 2009). Thus, this could be an explanation why a smaller difference between the conditions was observed for the dorsolateral prefrontal cortex, in comparison to the ventromedial prefrontal cortex.

In addition, from the previous research from Irani et al. (2007) on the feasibility of fNIRS to measure neural activity was concluded that fNIRS is a promising technique due to its abilities to measure brain activity over extended time periods in a non-invasive, portable, low-cost and safe manner. Despite the promising results of the approach, fNIRS is still an emerging technique which needs cross-validation efforts by utilizing other neuroimaging approaches, such as fMRI, to establish its

competence. The results obtained in this study suggest that fNIRS might be an adequate technique to measure brain activity. In previous research, by Plassmann and colleagues (2012), it became apparent that humans are in a way conditioned to develop a preference for certain brands and that this is correlated with activation in the dlPFC and vmPFC. This result has also been found in the current study.

Interestingly, the current study obtained an unpredicted result. A strongly negative beta was perceived on the ventrolateral prefrontal cortex, for both the positive and the negative condition (figure 4C & D). According to Nozari and Thompson-Schill (2016) this activity correlates with the perception and processing of words, as it reflects Broca’s area. The participants were in fact observing a brand logo, which mainly contained words and thus processing that word, while the fNIRS measurement was obtained. This finding supports the hypothesis that fNIRS is a reliable approach to measure brain activity, as it perceives neural correlates which seem to be logically present at the time of assessment of the data.

Even though the results from this research seem to be in line with previous research, no hard conclusions can be drawn from the results due to some limitations this research faced. Such as, having a small sample size. Thirty subjects were taken into account for the data analysis, while a larger sample size would have given a broader perspective on the research question. Nonetheless, this research can still contribute to the general knowledge and be considered in a meta-analysis, therefore effectively broadening the overall sample size.

In addition, another shortcoming in this research is that for the data analysis solely the oxygenated haemoglobin in the blood of the PFC was examined. However, the deoxygenated haemoglobin was not taken into consideration for the data analysis. The reason why in this research oxygenated haemoglobin was considered, is due to its role as an indicator for synaptic activity (Arthurs & Boniface, 2003). Though, as argued in the same study, the oxygenated haemoglobin seems to lack a representation for the neural activity of action potentials generated by neurons. Moreover, oxygenated haemoglobin seems to be comparable to the BOLD signal in fMRI (Carrion & Leon-Dominguez, 2012). In the same study, they argue that the oxygen metabolism rises with neural activity and the cerebral blood flow rises even higher. The oxygen metabolism results in little oxygenated haemoglobin and more deoxygenated haemoglobin, whereas the cerebral blood flow results in an eminent amount of oxygenated haemoglobin and a decreased amount of deoxygenated haemoglobin. Altogether, this results in a slightly larger amount of oxygenated haemoglobin in comparison to deoxygenated haemoglobin during neural activity. Arguing that oxygenated haemoglobin is a superior predictor for neural activity. However, previous research by Steinbrink et al. (2006), suggests that deoxygenated haemoglobin is more correlated to BOLD signal than oxygenated haemoglobin. Replication of this research, whilst observing the deoxygenated response, could therefore publish distinct results. Future research should investigate this further to draw decisive conclusions on this matter.

Furthermore, the experiment as performed in this study, could be viewed as Western Educated Industrialized Rich and Democratic (WEIRD) research. This is mainly due to the fact that it was conducted in The Netherlands, a WEIRD country. fNIRS, however, has the ability to be transported to areas everywhere in the world, including more rural areas. Because the most frequently used neuroscientific research methods, such as fMRI, are unable to be transported and most neuroscientific research is conducted in the western world. Making research less WEIRD has appeared to be a challenge. Due to the portability of fNIRS, research can be conducted to a greater extent of places on earth, therefore making research less WEIRD (Burns et al., 2019). Expanding

view of neuroscientific evidence. In addition, this might help people from outside the WEIRD countries.

An issue, illustrated in the introduction, is that fNIRS has a couple of disadvantages, among which the inability to measure the deep brain (Irani et al., 2007). However, a research group: Liu, Cui, Bryant, Glover and Reiss (2015), may have found a solution for one of these disadvantages. The research group managed to develop a support vector regression algorithm to predict deep brain activity derived from fNIRS results. They tested to what extent the predictions of the algorithm based on cortical activity, correlate with the BOLD response of the deep brain derived from fMRI. They found the highest correlation between the predictions of the algorithm for the deep brain from the cortex derived with fMRI and the fMRI derived deep brain response at the left insula, with a correlation coefficient of 0.90. The overall prediction performance has a mean of 0.67 across all regions and tasks. Due to these results, they conclude that deep brain activity can be predicted from cortical activity. The prediction of the algorithm derived from fNIRS measurements, show an overall correlation coefficient higher than 0.50, with at the top 15% a mean correlation of 0.70 or higher. These results are promising and have the potential to extend the competence of fNIRS to measure deep brain activity. Therefore, this algorithm could have great implications for the future of fNIRS capabilities.

To summarize, the results obtained in this study show that the dorsolateral prefrontal cortex and ventromedial prefrontal cortex are involved in observing a brand logo. Whereas, not the entire PFC is involved with brand perception. This is in line with previous research (e.g. Gilbert et al., 2010; Martin & Lawrence, 2003; Plassmann et al., 2012). These findings indicate that fNIRS is a promising approach to measure neural activity and can successfully distinguish the perception of positively and negatively defined brand logos. However, no hard conclusions can be drawn from these results, due to a few shortcomings in this experiment. Further research should replicate this experiment with a larger sample size to obtain a more representable dataset and draw conclusions. In addition, further research should explore the abilities to amplify fNIRS, to measure deep brain areas, made possible by a support vector regression algorithm. The utilization of fNIRS in neuroscientific and clinical research may have great implications for the future of neuroimaging techniques.

References

Acikalin, M. Y., Watson, K. K., Fitzsimons, G. J., & Platt, M. L. (2018). Rhesus macaques form preferences for brand logos through sex and social status based advertising. PloS one, 13, e0193055. doi: 10.1371/journal.pone.0193055

Arthurs, O. J., & Boniface, S. J. (2003). What aspect of the fMRI BOLD signal best reflects the underlying electrophysiology in human somatosensory cortex?. Clinical neurophysiology: official journal of the International Federation of Clinical Neurophysiology, 114, 1203–1209. doi: 10.1016/s1388-2457(03)00080-4

Brodmann, K. (1909). Vergleichende Lokalisationslehre der Grosshirnrinde in ihren Prinzipien dargestellt auf Grund des Zellenbaues. In J.A. Barth (Eds.), Leipzig.

Brodmann, K. (1912). Neue Ergebnisse über die vergleichende histologische Lokalisation der Grosshirnrinde. Anatomischer Anzeiger, 41, 157-216.

Burns, S. M., Barnes, L. N., McCulloh, I. A., Dagher, M. M., Falk, E. B., Storey, J. D., & Lieberman, M. D. (2019). Making social neuroscience less WEIRD: Using fNIRS to measure neural signatures of persuasive influence in a Middle East participant sample. Journal of Personality and Social Psychology, 116, 1–11. doi: 10.1037/pspa0000144

Camus, M., Halelamien, N., Plassmann, H., Shimojo, S., O’Doherty, J., Camerer, C., & Rangel, A. (2009). Repetitive transcranial magnetic stimulation over the right dorsolateral prefrontal cortex decreases valuations during food choices. European Journal of Neuroscience, 30, 1980-1988. doi: 10.1111/j.1460-9568.2009.06991.x

The Coca-Cola company. (2020). The Story of One of the Most Memorable Marketing Blunders Ever. Retrieved from https://www.coca-colacompany.com/news/the-story-of-one-of-the-most-memorable-marketing-blunders-ever.

D'esposito, M., Cooney, J., Gazzaley, A., Gibbs, S., & Postle, B. (2006). Is The Prefrontal Cortex Necessary for Delay Task Performance? Evidence from Lesion and fMRI Data. Journal Of The International Neuropsychological Society, 12, 248-260. doi: 10.1017/S1355617706060322 Ferrari, M., Mottola, L., & Quaresima, V. (2004). Principles, techniques, and limitations of near infrared spectroscopy. Can. J. Appl. Physiol., 29, 463-487. doi: 10.1139/h04-031

Gilbert, S.J., Henson, R.N.A., & Simons, J.S. (2010). The Scale of Functional Specialization within

Human Prefrontal Cortex. Journal of Neuroscience, 30,

1233-1237. doi: 10.1523/JNEUROSCI.3220-09.2010

Goel, V., & Dolan, R. (2001). The functional anatomy of humor: segregating cognitive and affective components. Nat Neuroscience, 4, 237–238. doi: 10.1038/85076

Gratton, G., Maier, J.S., Fabiani, M., Mantulin, W.W., & Gratton, E. (1994). Feasibility of intracranial near‐infrared optical scanning. Psychophysiology, 31, 211-215. doi: 10.1111/j.1469-8986.1994.tb01043.x

Hermans, D., De Houwer, J., & Eelen, P. (1994). The Affective Priming Effect: Automatic Activation Of Evaluative Information In Memory. Cognition And Emotion, 8, 515–533. doi:10.1080/02699939408408957.

Hosseini, H.S.M., Mano, Y., Rostami, M., Takahashi, M., Sugiura, M., & Kawashima, R. (2011). Decoding what one likes or dislikes from single-trial fNIRS measurements. NeuroReport, 22, 269-273. doi: 10.1097/WNR.0b013e3283451f8f

Huppert, T.J., Hoge, R.D., Diamond, S.G., Franceschini, M.A., & Boas, D.A. (2006). A temporal comparison of BOLD, ASL, and NIRS hemodynamic responses to motor stimuli in adult humans. NeuroImage, 29, 368-382. doi: 10.1016/j.neuroimage.2005.08.065.

International Commission on Non-Ionizing Radiation Protection (ICNIRP). (2006). Statement On Far

Infrared Radiation Exposure. Health Physics, 91, 630-645. doi:

10.1097/01.HP.0000240533.50224.65

Irani, F., Platek, S.M., Bunce, S., Ruocco, A.C., & Chute, D. (2007). Functional Near-Infrared Spectroscopy (fNIRS): An Emerging Neuroimaging Technology With Important Applications For The Study Of Brain Disorders. Clinical Neuropsychologist, 21, 9–37. doi: 10.1080/13854040600910018

Jöbsis F. F. (1977). Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science, 198, 1264–1267. doi: 10.1126/science.929199

Klucharev, V., Smidts, A., & Fernández, G. (2008). Brain mechanisms of persuasion: how ‘expert power’ modulates memory and attitudes. Social Cognitive and Affective Neuroscience, 3, 353– 366. doi: 10.1093/scan/nsn022

Koenigs, M., & Tranel, D. (2008). Prefrontal cortex damage abolishes brand-cued changes in cola preference. Social Cognitive and Affective Neuroscience, 3, 1–6. doi: 10.1093/scan/nsm032 Leon-Carrion, J., & Leon-Dominguez, U. (2012). Functional Near-Infrared Spectroscopy (fNIRS): Principles and Neuroscientific Applications. In P. Bright (Eds.), Neuroimaging: Methods, 47-74. In-Tech.

Liu, N., Cui, X., Bryant, D., Glover, G., & Reiss, A. (2015). Inferring deep-brain activity from cortical activity using functional near-infrared spectroscopy. Biomed. Opt. Express, 6, 1074-1089. doi: 10.1364/BOE.6.001074

Martin, P.P., & Lawrence, F.R. (2003). Ventromedial prefrontal cortex activation is critical for

preference judgments. NeuroReport, 14, 1311-1315. doi:

10.1097/01.wnr.0000078543.07662.02

Nozari, N., & Thompson-Schill, S.L. (2016). Chapter 46: Left Ventrolateral Prefrontal Cortex in Processing of Words and Sentences. In G. Hickok & S.L. Small (Eds.), Neurobiology of language, 569-584. Academic Press. doi: 10.1016/B978-0-12-407794-2.00046-8.

Plassmann, H., Ramsøy, T.Z., & Milosavljevic, M. (2012). Branding the brain: A critical review and outlook. Journal of Consumer Psychology, 22, 18-36. doi: 10.1016/j.jcps.2011.11.010

Schaefer,M., Berens, H., Heinze, H.J., & Rotte, M. (2006). Neural correlates of culturally familiar brands of car manufacturers. NeuroImage, 31, 861-865. doi: 10.1016/j.neuroimage.2005.12.047 Steinbrink, J., Villringer, A., Kempf, F., Haux, D., Boden, S., & Obrig, H. (2006). Illuminating the BOLD signal: combined fMRI–fNIRS studies. Magnetic resonance imaging, 24, 495-505. doi: 10.1016/j.mri.2005.12.034

Appendix

A. Lay-out of the fNIRS optodes

B. fNIRS channels FNIRS CHANNEL LOCATION 1 Rx1 - Tx1 HHb 2 Rx1 - Tx1 O2Hb 3 Rx1 - Tx2 HHb 4 Rx1 - Tx2 O2Hb 5 Rx1 - Tx3 HHb 6 Rx1 - Tx3 O2Hb 7 Rx2 - Tx1 HHb 8 Rx2 - Tx1 O2Hb 9 Rx2 - Tx3 HHb 10 Rx2 - Tx3 O2Hb 11 Rx2 - Tx4 HHb 12 Rx2 - Tx4 O2Hb 13 Rx3 - Tx2 HHb 14 Rx3 - Tx2 O2Hb 15 Rx3 - Tx3 HHb 16 Rx3 - Tx3 O2Hb 17 Rx4 - Tx3 HHb 18 Rx4 - Tx3 O2Hb 19 Rx4 - Tx4 HHb 20 Rx4 - Tx4 O2Hb 21 Rx3 - Tx5 Hhb 22 Rx3 - Tx5 O2Hb 23 Rx4 - Tx5 HHb 24 Rx4 - Tx5 O2Hb 25 Rx5 - Tx7 HHb 26 Rx5 - Tx7 O2Hb 27 Rx5 - Tx6 HHb 28 Rx5 - Tx6 O2Hb 29 Rx5 - Tx8 HHb 30 Rx5 - Tx8 O2Hb 31 Rx7 - Tx7 HHb 32 Rx7 - Tx7 O2Hb 33 Rx7 - Tx8 HHb 34 Rx7 - Tx8 O2Hb 35 Rx7 - Tx10 HHb 36 Rx7 - Tx10 O2Hb 37 Rx6 - Tx6 HHb 38 Rx6 - Tx6 O2Hb 39 Rx6 - Tx8 HHb 40 Rx6 - Tx8 O2Hb 41 Rx8 - Tx8 HHb 42 Rx8 - Tx8 O2Hb 43 Rx8 - Tx10 HHb 44 Rx8 - Tx10 O2Hb 45 Rx6 - Tx9 HHb 46 Rx6 - Tx9 O2Hb 47 Rx8 - Tx9 HHb 48 Rx8 - Tx9 O2Hb C. Questionnaire questions

(Wanneer u een bedrijf niet kent of geen mening heeft over een bedrijf, hoeft u deze niet in een categorie te plaatsen)

2. Selecteer de bedrijven waar u juist wel/juist niet producten van zou kopen.

(Wanneer u een bedrijf niet kent of geen mening heeft over een bedrijf, hoeft u deze niet in een categorie te plaatsen)

3. Selecteer de bedrijven die u juist wel/juist niet zou aanbevelen aan een bekende. (Wanneer u een bedrijf niet kent of geen mening heeft over een bedrijf, hoeft u deze niet in een categorie te plaatsen)

4. Selecteer de bedrijven die u juist wel/juist niet 'goed' voor de wereld acht, met goed voor de wereld wordt bedoelt: dit (bedrijf) levert een positieve bijdrage aan de wereld. (Wanneer u een bedrijf niet kent of geen mening heeft over een bedrijf, hoeft u deze niet in een categorie te plaatsen)

D. Stimuli: Brand logos Version 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

16 17 18 19 20 21 22 23 24 25 Version 2 26 27 28 29 30 31 32 33 34 35 36 37 38

39 40 41 42 43 44 45 46 47 48 49 50 Version 3 51 52 53 54 55 56 57 58 59 60 61 62 63

64 65 66 67 68 69 70 71 72 73 74 75 Version 4 76 77 78 79 80 81 82 83 84 85 86 87 88

89 90 91 92 93 94 95 96 97 98 99 100 Version 5 101 102 103 104 105 106 107 108 109 110 111 112 113

114 115 116 117 118 119 120 121 122 123 124 125 Version 6 126 127 128 129 130 131 132 133 134 135 136 137 138

139 140 141 142 143 144 145 146 147 148 149 150