Keywords

1 Introduction

With the massive information and service content which tend to be homogenized filled in our daily life, the competitive pressures on websites are unprecedented. In order to help users to recognize brand and cultivate brand loyalty, the interactive system has increasingly become user’s main way in brand recognition. Brand recognition will affect the user’s purchase decisions and brand loyalty to a great extent. Howard pointed out that consumers’ cognition of the brand and product is from the product information, which will be transformed into brand attitudes and will further influence the purchase intention and purchase behavior of the brand [1]. As a result, we now know that users tend to choose the product which has higher brand recognition when they do the purchase decisions. If the location of brand information appears more likely to be noticed, it can deepen user’s cognition for brand and then promote user’s brand recognition. As a result, website which has higher recognition can be easily distinguished from other websites and can occupy psychological position of users.

In recent years, many researchers have conducted brand recognition as follows: Kapferer suggested that the market had experienced the changes of brand image, brand positioning and brand personality, then an era of brand recognition will come. He put forward a hexagon prism model and brand pyramid theory in the book Strategic Brand Management [2]. Schniitt and Simonson proposed a brand recognition model from marketing aesthetics perspective. They considered brand recognition included four elements: properties, products, presentations and publications [3]. Aaker argued that brand identification should include soul of brand, core identity of brand and extended identity of brand in the book Brand Leadership [4]. Kotler studied brand recognition from the point of marketing communication and considered that brand was the means of communication. The information that company conveyed to consumers by brand was three-dimensional and all-around. It also included three-layer construction, namely, brand recognition, brand Image and brand personality [5]. Aaker proposed a ‘four-level theory’, he proposed that brand identification was composed of 12 elements in four levels: product identification (product area, product properties, quality value, usage, user, country of origin), brand organization (organizational nature, localization or globalization), personal brand (brand personality, relationship between the brand and customer) and symbolic brand (visual image/metaphor and brand tradition) [6]. Maria found the relationship among brand identification, brand image and brand preference in the network environment. He also verified that good brand recognition was beneficial to brand image and brand preference. Furthermore, offline influence had greater significance than online influence [7]. Bernd constructed a consumer psychology model through the integration of relevant research results, showing the consumer’s brand awareness, judgments and related processes [8]. da Silveira et al. reconstructed the brand recognition model under the influence of many factors, such as time, consumer and so on [9].

The websites have the characteristic of content pre-guiding. Navigation bar which is the catalogue of the website content and has a higher utilization may directly influence user experience. Furthermore, users usually give priority concerns to the navigation bar when they enter the home page. In addition, the position of the navigation bar is fixed on the top of the page, which may help user to know where they are and can also play an important role in brand recognition. Therefore, this paper chooses the elements in navigation bar for brand recognition research.

At present, there are few researches on LOGO location in navigation bar. However, there are many papers on graphic and text design in LOGO. Li studied the elements of corporate culture, physical geography and environmental characteristics, the traditional cultural characteristics and the corresponding graphics and text in the hotel visual image LOGO design [10]. Wang argued that how to improve brand recognition from the perspective of LOGO attribute recognition and LOGO recognition in same class [11]. Niu discussed the importance of LOGO design in the brand and different design forms in LOGO design [12]. Xu found that lowercase wordmarks can increase perception of brand friendliness compared with the uppercase wordmarks. In addition, consumers felt that uppercase wordmarks were more authoritative [13]. Liu found the approach to combine traditional Chinese art elements with LOGO design [14]. Therefore, considering brand LOGO itself is not enough, no matter how good the brand is, if it cannot be recognized by the user, it will be nothing. If the application scenario of LOGO can be taken into account, in some way to enhance the user’s brand cognition, it will attract more users and improve brand recognition.

In the past, the main method of brand recognition were subjective evaluation and physiological evaluation. The former obtains the brand recognition from users through questionnaire, user interview, etc. The latter is mainly used to obtain the physiological indexes to understand the degree of brand recognition from users, such as Electromyography (EMG), eye movement and so on. However, it still has some shortcomings: for subjective evaluation, it may lack reliable and repeatable objective indicators to evaluate; for physiological evaluation, it is easy to be disturbed by external factors. Event-related Potential (ERPs) uses brain-imaging techniques for positive and negative waves of the scalp-related potential in order to reflect the subtle brain cognitive activity. Moreover, it can describe and analyze the cognition process of digital interface, which directly reflect neuroelectricity activity and establishing the relationship between neurophysiology and cognitive process.

In the visual Oddball paradigm, the user’s recognition process of target stimulus includes working memory, attention, stimulation evaluation and pattern matching, etc. The appearance of target stimulus may induce more significant effect on P300 component [15]. Hillyard proposed that during the selective attention process, P3 represented the response set, which represented the subsequent stage of the perceived information processing, and later it was confirmed that a larger P300 amplitude may appear when people pay more attention to something [16]. Donchin pointed out that the latency of the P300 reflected the speed and selection process of stimulus recognition when the memory was updated. In addition, he also believed that the content which induced higher amplitude of P300 was more likely to be remembered [17].

In this paper, the opinion that the brand recognition is not only dependent on the internal factors of the brand is put forward. From a new perspective, LOGO Location in the navigation bar is an important external factor that affects the web brand recognition, which can provide a more objective method for brand building. Oddball experimental paradigm is used to explore user’s EEG psychological characteristics induced by the LOGO’s different locations in the navigation bar. With the help of P300 amplitude changes and subjective evaluation results, the effect of LOGO location in navigation bar on web brand recognition can be revealed. If the brand appears more likely to be noticed, it can deepen user’s cognition for brand and then promote user’s brand recognition. What’s more, it may help users recognize and choose websites rapidly according to LOGO location in navigation bar, enhancing the brand recognition and promoting brand loyalty of the website that have important influence on website building.

2 Experiment Design

2.1 Participants

Participants consist of 20 students who are between 22 and 27 years old with a ratio of one to one, right-handed. They all have corrected visual acuity over 1.0 and have no history of mental illness.

2.2 Experiment Method

ERP experiment

Experiment Materials.

Oddball experimental paradigm is used in the experiment. According to the different LOGO locations, three sets of location relationship are established. Plan A places the LOGO on the left (Fig. 1), Plan B places the LOGO on the middle (Fig. 2), Plan C places the LOGO on the right (Fig. 3). Plan A, Plan B, Plan C are target stimuli, Fig. 4 is the standard stimulus. It is predicted that the target stimulus will induce significant P300 in the Oddball experimental paradigm.

Fig. 1.
figure 1

LOGO placed on the left.

Full size image
Fig. 2.
figure 2

LOGO placed on the middle.

Full size image
Fig. 3.
figure 3

LOGO placed on the right.

Full size image
Fig. 4.
figure 4

Standard stimulus (no LOGO).

Full size image

Experimental Procedure.

In order to avoid the changes in the amplitude of P300 caused by the probability variation of the target stimulus, the number of target stimuli and the standard stimuli are randomly distributed in the four experimental modules. In order to eliminate the interference of color, icon and other related factors, all the stimuli images are desaturated and placed in the center of the screen. The background color is white and the stimulus is black. The stimulus is in the 17-in. display. Participants were asked to sit in front of the display screen with a distance of 55–60 cm, facing the computer screen. The experimental program was written by E-PRIME, and the behavioral data were collected. At the beginning of the experiment, the computer screen showed experimental guidance. After reading the guidance, participants started the experiment by pressing any key. The experiment was divided into two parts: practice period and formal experiment period. The cross-visual guidance appeared in the center of screen for 500 ms at first, then standard stimulus and target stimulus appeared at random. Azizian et al. set the stimulus presentation time to 500 ms to eliminate visual residuals [18]. The participants were told to respond at the fastest speed and press ‘L’ to respond to target stimulus. Throughout the testing process, the standard stimuli appeared 480 times in total (the number of standard stimuli account for 80%), each kind of target stimuli appeared 40 times (target stimuli account for 20%). The experiment procedure is shown below (Fig.  5).

Fig. 5.
figure 5

Experiment procedure.

Full size image

EEG Recording and Data Analysis.

The experiment data collection and data analysis are conducted by BP EEG recording system. The brain waves induced by target stimuli are recorded by using EEG Recording Cap Set for 32 Channels (0.05–100 Hz band pass; 1 kHz/channel sampling rate). Electrode impedance is maintained below 5 kΩ throughout the experiment. According to the literature description and the total waveform of P300, the regions that have the highest brain activation between 0 ms and 800 ms are centro-parietal lobe, parietal lobe and occipital lobe [19], so CP5, CP1, CP2, CP6, P7, P3, Pz, P4, P8, POz, O1, Oz, O2 are selected (see Fig. 6). The amplitude and latency are compared by analysis of variance via SPSS. The p value of variance analysis is calibrated with the Greenhouse Geisser method.

Fig. 6.
figure 6

Electrode diagram

Full size image

2.3 Subjective Evaluation

After ERP experiment, participants were allowed to finish a questionnaire (see Fig. 7). The effect of LOGO location in navigation bar on web brand recognition was evaluated by 7-level Likert scale to verify the reliability and effectiveness of ERP.

Fig. 7.
figure 7

Questionnaire about brand recognition

Full size image

3 Experimental Results and Analysis

3.1 Subjective Evaluation Data

According to statistics, LOGO on the left was rated 5.36 points on average among 22 subjects; LOGO on the middle was rated 5.59 points; LOGO on the right was rated 3.37 points. Through further analysis of the results, we found that 57% subjects thought the LOGO placed on the middle had the highest brand recognition; 34% subjects thought the LOGO placed on the left had the highest brand recognition; only 9% subjects hold the view that the LOGO placed on the right had the highest brand recognition. Based on the subjective evaluation results, this paper will explore the relationship between LOGO location and brand recognition further.

3.2 Behavioral Data

Table 1 showed three groups of behavioral data, collecting 22 valid samples in total. From the table, it can be safely concluded that the mean reaction time in condition 1 (LOGO on the left) was longer than condition 2 (LOGO on the middle) and condition 3 (LOGO on the right), and these three groups of mean reaction time had little dispersion degree. The accuracy of three experimental groups were more than 98%, which indicated that the subjects in the three experimental groups maintained high accuracy. Mean reaction time and accuracy in three experimental groups had little difference, so in-depth interactive effect analysis was not conducted.

Table 1. Descriptive statistics of behavioral data in 3 groups of experiments.
Full size table

3.3 EEG Data Analysis

Figure 8 showed the EEG waveforms of 13 electrodes CP5, CP1, CP2, CP6, P7, P3, Pz, P4, P8, POz, O1, Oz and O2 from 0 ms to 500 ms. The P2 and P3 components were all induced under the three experimental conditions (LOGO placed on the left, middle and right). The attention, repetition and stimulation may affect the amplitude of P2. This paper analyzed the P3 that appeared in 200–500 ms.

Fig. 8.
figure 8figure 8

Under condition of different types total mean ERP waveforms

Full size image

P300 Analysis.

Tables 2, 3 and 4 showed the mean amplitude of CP5, CP1, CP2, CP6, P7, P3, Pz, P4, P8, POz, O1, Oz and O2 which appeared 200 ms to 500 ms after the target stimuli. With respect to P300 component, repeated ANOVA of mean amplitude for 200–500 ms in 3 (LOGO position: LOGO placed on the left, middle, right)) × 13 (electrode position: CP5, CP1, CP2, CP6, P7, P3, Pz, P4, P8, POz, O1, Oz, O2) was conducted. The statistical analysis results were shown in Table 5: (1) The main effect of the LOGO location was significant (F = 5.938, p = 0.005 < 0.01); The mean amplitude of LOGO placed on the right (3.59 μν) was slightly larger than LOGO placed on the left (3.48 μν); The mean amplitude of LOGO placed on the left/right was significantly larger than LOGO placed on the middle (2.58 μν); (2) The main effect of electrode position was marginal significant (F = 2.826, P = 0.052 > 0.05); (3) The interaction effect between LOGO location and electrode position was significant (F = 8.727, P = 0.000 < 0.01).

Table 2. The mean amplitude of P300 at different electrode positions (part 1).
Full size table
Table 3. The mean amplitude of P300 at different electrode positions (part 2).
Full size table
Table 4. The mean amplitude of P300 at different electrode positions (part 3).
Full size table
Table 5. The results of variance analysis for two factors (mean amplitudes).
Full size table

In order to explore the latency differences between the target and standard stimulus further, repeated ANOVA of mean latency in 3 (LOGO position: LOGO placed on the left, middle, right) × 13 (electrode position: CP5, CP1, CP2, CP6, P7, P3, Pz, P4, P8, POz, O1, Oz, O2) was conducted. The results were shown in Table 6 as follows: (1) The main effect of the LOGO location was significant (F = 6.373, p = 0.009 < 0.01); The latency of LOGO placed on the left (308 ms) < LOGO placed on the right (315 ms) < LOGO placed on the middle (331 ms); (2) The main effect of electrode position was significant (F = 3.613, P = 0.005 < 0.01); (3) The interaction effect between LOGO location and electrode position was significant (F = 5.242, P = 0.000 < 0.01).

Table 6. The results of variance analysis for two factors (mean latency).
Full size table

3.4 EEG Mapping Analysis

As shown from Fig. 9, the brain activation regions induced by LOGO placed on the left appeared in the left hemisphere and occipital lobe; The brain activation region induced by LOGO placed on the middle mainly appeared in the occipital lobe; The brain activation region induced by LOGO placed on the right appeared in the right hemisphere and occipital lobe. According to the forementioned, brain activation region induced by LOGO placed on the right was the largest, brain activation region induced by LOGO placed on the left took second place and brain activation region induced by LOGO placed on the middle was the smallest. As time goes on, brain activation shifted to the front of brain, indicating that the brain had a deeper cognitive process for stimulation.

Fig. 9.
figure 9

Under condition of different types EEG Mapping

Full size image

4 Discussion

In this experiment, the LOGO was placed on the left/middle/right under three different experimental conditions, all the target stimuli can induce significant P300. Donchin et al. considered that the P300 was related to the cognitive schema update. When the target stimulus deviated from the standard stimulus which stored in working memory, the P300 can be induced [17]. After analyzing the mean amplitude of p300 induced by the different LOGO locations, LOGO placed on the right can generate the largest p300 amplitude, LOGO placed on the left takes second place, LOGO placed on the middle generates the smallest P300 amplitude. Additionally, there was significant difference between LOGO placed on the left/right and middle, indicating that LOGO placed on the left/right can attract more attention than LOGO placed on the middle. The mean amplitude difference of P300 in the three experiments may be related to the brand recognition. The amplitude of P300 was proportional to the amount of psychological resources which users had invested in. The more attention resources users invested in the recognition of target stimuli from standard stimuli, the larger the amplitude of the P300 was. The amplitude of P300 induced by target stimulus was proportional to brand recognition. The P300 appeared 300 ms to 600 ms after the target stimulus. In this paper, the mean latency of P300 is about 318 ms. Donchin et al. believed that the latency of P300 was only influenced by stimulus discriminability [20]. In this experiment, P300 induced by the user’s recognition of the target stimulus was slightly delayed. After the latency of P300 was compared among three groups, the results showed that the experimental conditions had significant effect on the latency of P300 and the mean latency of LOGO placed on the left/right side was significantly shorter than the mean latency of the LOGO placed on the middle, indicating that LOGO placed on the left/right was conductive to brand recognition in navigation bar.

5 Conclusion

This paper first proposed the view that website brand recognition is influenced not only by the LOGO’s image of the brand, but also its location in the web navigation bar. By analyzing the P300 amplitude of the LOGO located in three certain positions in the bar, namely on the left, on the middle and on the right side separately, result shows that the LOGO placed on the left/right can generate the largest p300 amplitude, occupying the most user’s attention resources. This can further deepen users’ awareness of the brand, thereby enhancing the website brand recognition as well as users’ loyalty to the brand, which ultimately determines users’ purchase decisions. Therefore, the company can promote customers’ brand recognition with the ERPs technology in order to obtain better resources, to achieve a dominant position in market competitions, to erect barriers over the competitors, and finally to stand out among similar websites. Considering that this paper has only considered the impacts that LOGO location in the navigation bar on users’ brand recognition, follow-up studies will look into more other factors of the websites, including the colors, the styles and so forth in the future.