Mathematics anxiety, an emotional state resulting in negative responses to math problems and numerical information, has been extensively studied in educational psychology. Research on the impact of math anxiety on consumer purchase decisions, however, is still in its nascent stages. This paper examines the interaction between math anxiety and different promotion framing formats on consumers’ perceived savings, price acceptability, and purchase decisions. Across two studies, we demonstrate that consumers with varying levels of math anxiety respond differently to various promotion frames: a gain versus a reduced loss and a single discount versus multiple discounts. We further show that consumers with insufficient math ability may experience negative affect and heighten math anxiety, particularly when faced with numerical and arithmetic tasks commonly encountered while shopping.
Notes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
1 Introduction
Mathematics anxiety, or the fear of having to do mathematical computations, is a highly discussed issue in educational psychology. Students with high math anxiety tend to escape math classes, avoid arithmetic tasks, and achieve lower grades in math tests (Ashcraft & Kirk, 2001). Math anxiety may also interfere with consumers’ processing of numerical information and subsequently affect their price perceptions and judgments. The case of A&W burgers, for example, vividly shows how consumers’ inability to understand fractional magnitudes affected their purchase decisions. Consumers misperceived A&W’s 1/3-pound burger as smaller than McDonald’s 1/4-pound burger since “3” is smaller than “4.” They thus chose the more expensive McDonald’s burger, even though they would have preferred the taste of A&W’s burger instead. Consumers’ math anxiety in this scenario resulted in the increasing use of heuristics, relying on partial information, avoiding calculations, and making overall suboptimal or incorrect judgments (i.e., since 4 is greater than 3, 1/4 must be bigger than 1/3) (Green, 2014).
While the impact of math anxiety on academic performance has been extensively studied, research into its influence on consumer purchase behavior is relatively underexplored. Marketers employ various pricing tactics, such as price promotions, to persuade consumers to buy their products and services. It is essential for marketers to understand how math anxiety affects consumers’ product preferences across different price points and promotional formats, as effective price promotions can positively influence buyers’ perceptions of deal savings and subsequent purchase decisions (Chen et al., 1998). There are many ways to present (i.e., frame) price promotion, with each frame affecting consumers’ evaluations of a deal offer differently (Krishna et al., 2002). Consumers’ fluency in processing price information further influences their preferences for different price promotion formats (Coulter & Roggeveen, 2014). If consumers do not perceive a deal or an increase in value, the promotion may not have the desired effect on the marketers’ revenues. Research in cognitive arithmetic suggests that when people face difficult math problems, the math anxiety that they experience inhibits their math performance by reducing their working memory capacity when conducting the calculations (Ashcraft & Kirk, 2001). Consumers afflicted with math anxiety may avoid numerical calculations and choose an alternative instead if a price promotion frame requires computations to determine the final price to be paid (Suri et al., 2013).
Advertisement
In this paper, we investigate the influence of different price promotion frames on the relationship between math anxiety and price perceptions, resulting in differences in purchase intentions. This research contributes to the pricing and marketing literature by bridging the gap between the findings of education research regarding the impact of math anxiety on academic performance and the practical implications of math anxiety on consumers’ price perceptions. Furthermore, we identify certain types of promotion frames that facilitate math-anxious consumers to perceive greater savings and price acceptability, thereby increasing their willingness to purchase.
2 Theoretical framework
2.1 Math anxiety and price computation
Math anxiety is an apprehension of mathematics ranging from feelings of discomfort with numerical operations to avoiding math classes (Mathison, 1977). It may reduce the capacity of working memory for arithmetic, resulting in a longer response time, less accuracy, and more errors when processing numerical data in individuals with high math anxiety. These individuals exhibit negative affective reactions to scenarios requiring numbers, math, and arithmetic calculations (Ashcraft & Moore, 2009). Affect refers to a set of interrelated emotional states that can be positive or negative (Russell, 1980). Recent neuroscience research indicates that both math anxiety and negative emotions originate in the prefrontal cortex of the brain (Diekhof et al., 2011). People who experience numerical anxiety and negative affect may avoid math-related situations or make inaccurate estimations.
Cognitive psychology has extensively studied math anxiety among school-aged children and explored the relationship between math anxiety and math ability. Math ability, which includes an individual’s proficiency in math skills and knowledge of arithmetic rules to solve math problems, plays a crucial role in enhancing math performance (Ashcraft & Kirk, 2001; Suri et al., 2013). Previous research suggests that young children, such as first-graders, experience mild math anxiety, with its intensity potentially increasing over time, particularly due to prior poor math performance or difficulties in different math skills (Aarnos & Perkkilä, 2012; Szczygieł & Pieronkiewicz, 2022). We predict that insufficient math ability could intensify math anxiety, which may persist into adulthood. The lack of sufficient math skills may develop a negative affect towards numbers and arithmetic problems, thereby increasing math anxiety.
H1: Consumers with low math ability will experience negative affect and higher math anxiety when faced with numerical and arithmetic tasks.
Advertisement
2.2 Math anxiety, price perception, and promotion framing
Consumers generally seek to minimize their monetary sacrifice and maximize savings when making purchase decisions. When they find deals where the promotional price is comparable to or lower than their expectations, they may perceive it as fair, reasonable, and acceptable, thus increasing the likelihood of making a purchase (Xia & Monroe, 2004). Nevertheless, people often make choices based on affective responses rather than cognitive deliberations (Laros & Steenkamp, 2005). Consumers with high math anxiety may feel dissatisfied with a price promotion, especially if it involves complex formats or requires computations to calculate the final price or savings. Consequently, they may perceive the promotional price as unfair, unreliable, or nontransparent (Matzler et al., 2007), leading to greater uncertainty regarding product choice and reducing their purchase intention (Suri et al., 2013).
Consumers’ reactions to price promotions are influenced by their interpretation of the numerical values, the perceived price difference, and the words (semantic cues) used in the promotional message (Monroe, 2022). The relative ease and motivation for processing the numerical and verbal information determine these perceptions and interpretations. When promotions are presented in relative terms, such as a percentage off the original price, consumers typically need to calculate the savings and final selling price. In contrast, promotions offered in nonmonetary units, such as free gifts or bonus packs, may be perceived as separate gains since the promotion is in a different metric from the merchandise price. The benefits of such a promotion to buyers are comparatively more difficult to integrate into the final price, leading consumers to perceive them as a separate additional gain (Diamond, 1992).
Math-anxious consumers tend to avoid complex price computations and could prefer straightforward promotions such as free product offers, where the amount of savings is readily apparent, over more complex price promotions such as percentage-off deals. Conversely, consumers with low math-anxiety typically have less difficulty with price calculations and may therefore prefer price discounts that directly reduce their monetary sacrifice.
H2a: When a promotion is framed as buy-one-get-one-free, compared to a percentage-off, consumers with high math-anxiety will perceive greater savings, price acceptability, and purchase intention.
H2b: When a promotion is framed as percentage-off, compared to a buy-one-get-one-free, consumers with low math-anxiety will perceive greater savings, price acceptability, and purchase intention.
2.3 Math anxiety and discount framing
Presenting multiple discounts may enhance the perceived depth of the discount, savings, and overall value compared to a single discount of equivalent economic value when consumers simply add up the discounts (Chen & Rao, 2007). For instance, if a product is discounted by 25% and then by an additional 10%, buyers may perceive the total discount as 35%, resulting in greater savings than a single 33% discount. However, some consumers may perceive multiple discounts as providing less savings due to two heuristics. First, the anchoring-and-adjustment heuristic states that buyers perform only a few calculations while the anchor influences their purchase price estimation. They may reach a plausible estimate, but their adjustments are insufficient (Epley & Gilovich, 2006). In the scenario of multiple discounts of 25% off and an additional 10% off, buyers may estimate the final discount to be close to 25%. Second, the numerosity cue heuristic suggests that consumers who lack the cognitive resources or motivation to process price information value the size of the discount rather than the number of discounts, leading them to perceive multiple discounts as separate units and consequently less than a single discount (Cai & Suri, 2009).
Math anxiety can be a main reason for consumers to rely on these heuristics and shortcuts. Multiple discounts pose challenges in processing numerical information, leading individuals to favor simpler calculations (Coulter & Roggeveen, 2014). Conversely, low math-anxiety buyers may process price information with more ease and may even find multiple discounts more appealing than a single discount (Guha et al., 2018). Given their ability to compute the actual savings and make rational choices, whether a promotion offers a single or multiple discounts may not be a deal-breaker for them.
H3a: When a promotion is framed as a single discount, compared to multiple discounts, consumers with high math-anxiety will perceive greater savings, price acceptability, and purchase intention.
H3b: When a promotion is framed as a single discount, compared to multiple discounts, consumers with low math-anxiety will perceive no difference in savings, price acceptability, and purchase intention.
3 Study 1
3.1 Method
A 2 (math anxiety levels: low vs. high) × 2 (promotion frames: buy-one-get-one-free vs. percentage-off discount) between-subjects factorial experiment was conducted to test H1 and H2. An online video game was selected as the product based on pretest results. Data were collected from students at a public university (n = 184, 46% female), predominantly under 35 years old (86%). Cell sizes ranged from 38 to 54. Participants were shown a short scenario involving the purchase of two PlayStation video games. Promotion frames were manipulated through a product webpage with the product image, name, price, product specifications, and description. All participants were randomly assigned to the same product webpage but with different promotion frames (buy one at $56.69 and get the next one free vs. buy one at $37.79 and get the next one 50% off). The net price after promotion in both scenarios was the same at $56.69 (see Table 1).
Table 1
Experimental design of Studies 1 and 2
Promotion framing format
Subject paid (or saved)
Study 1: promotion frames
High vs. low math anxiety
Buy-one-get-one-free
$56.69+$0 = $56.69
Buy-one-get-next one-50% off
$37.79+($37.79*50%) = $56.69
Study 2: discount frames
High vs. low math anxiety
Single discount
$479.99*(1−25%) = $359.99
Multiple discounts
($479.99*(1−10%) − $32.00)*(1−10%) = $359.99
After browsing the webpage, participants responded to a 5-point Likert scale measuring negative affect (NA) and three 7-point Likert scales measuring perceived savings (PS), price acceptability (PA), and purchase intention (PI). To measure the level of math anxiety (MA), 1 week after the experiment, participants were asked to complete a 30-item MARS-Brief survey (Suinn & Winston, 2003). Each respondent self-reported his or her degree of anxiety concerning each math-related item on a 5-point Likert scale. Following an exploratory factor analysis, we retained 24 items on two dimensions, including math test anxiety and numerical task anxiety, aligning with Wilder (2013) findings. However, only 12 items of numerical task anxiety were included in the analysis as they are the most relevant measures of consumer math anxiety during purchasing situations. Additionally, participants completed a 20-item math ability test from the Kit of Factor-Referenced Cognitive Tests that measured arithmetic and mathematical aptitude (Ekstrom et al., 1976). The Appendix listed all measures, their sources, construct reliability, and average variance extracted (AVEs) calculated for both studies.
3.2 Results
Participants were divided into two high and low math-anxiety groups based on the sample mean math anxiety scores (NHigh = 81, NLow = 103). The high and low math-anxiety groups had significantly different levels of math anxiety (MHigh = 3.27, MLow = 1.40; F = 451.62, p < .001). To examine the difficulty of price computation in each promotion frame, participants estimated the final price for the two video games. Correct responses differed significantly between the two frames (MBuy-one-get-one-free = .53, MPercentage-off = .37; F = 5.02, p < .05), indicating higher difficulty with the percentage-off frame. In the buy-one-get-one-free condition, 63% of low math-anxiety participants correctly calculated the final price, compared to just 42% of those with high math-anxiety. In the percentage-off condition, only 46% of low math-anxiety respondents and 24% of high math-anxiety respondents provided accurate responses (see Table 2).
Table 2
Final price estimation accuracy in Studies 1 and 2
Low math anxiety
High math anxiety
Total sample
Total
Correct
Wrong
Total
Correct
Wrong
Total
Correct
Wrong
Study 1
Total sample
103
56
47
81
27
54
184
83
101
54%
46%
33%
67%
45%
55%
Buy-1-get-1-free
49
31
18
43
18
25
92
49
43
63%
37%
42%
58%
53%
47%
Percentage-off
54
25
29
38
9
29
92
34
58
46%
54%
24%
76%
37%
63%
Study 2
Total sample
81
28
53
155
42
113
236
70
166
35%
65%
27%
73%
30%
70%
Single discount
33
14
19
85
34
51
118
48
70
42%
58%
40%
60%
41%
59%
Multiple discounts
48
14
34
70
8
62
118
22
96
29%
71%
11%
89%
19%
81%
The bootstrapping method with 5000 samples and a 95% confidence interval (Hayes, 2017) was used to examine the mediation effect of negative affect in the relationship between math ability and math anxiety (H1). We standardized the factors in the model and mean-centered math ability. The results showed a significant effect of math ability on math anxiety through negative affect (β = −.32, t = −4.61, p < .001). Math ability adversely influenced negative affect (β = −.23, t = −3.26, p < .01), while negative affect positively impacted math anxiety (β = .35, t = 5.18, p < .001). When the mediator was included in the model, the direct effect of math ability on math anxiety was reduced but remained significant (β = −.24, t = −3.58, p > .001), suggesting a partial mediation effect. The indirect effect was significant (β = −.08, t = −2.93, p < .01, CI = −.14, −.03), with the mediator accounted for one-fourth of the total effect (PM = .25). The Sobel test confirmed a significant mediation effect (z = −4.26, p < .001), supporting H1.
A 2 (math anxiety levels: low vs. high) × 2 (promotion frames: buy-one-get-one-free vs. percentage-off discount) MANCOVA was conducted to examine the interaction effect on consumers’ price perception and purchase intention (H2). The final price calculation served as a covariate but was found insignificant (F(4, 179) = 1.04, p > .05). The MANOVA results showed a significant interaction (λ = .89; F(4, 179) = 5.43, p < .001). Subsequent ANOVA revealed a math anxiety × promotion frame interaction on perceived savings (F(4, 179) = 19.16, p < .001), price acceptability (F(4, 179) = 12.28, p < .01), and purchase intention (F(4, 179) = 10.87, p < .01).
When presented with a buy-one-get-one-free promotion, compared to a percentage-off discount, highly math-anxious participants perceived significantly greater savings (MBuy-one-get-one-free = 5.30, MPercentage-off = 4.65; F(2, 79) = 5.45, p < .05), price acceptability (MBuy-one-get-one-free = 5.33, MPercentage-off = 4.75; F(2, 79) = 4.16, p < .05), and purchase intention (MBuy-one-get-one-free = 4.74, MPercentage-off = 3.96; F(2, 79) = 4.64, p < .05), supporting H2a. Conversely, when offered a percentage-off discount, participants with low math-anxiety perceived substantially greater savings (MPercentage-off = 4.87, MBuy-one-get-one-free = 3.91; F(2, 99) = 13.23, p < .001), price acceptability (MPercentage-off = 5.39, MBuy-one-get-one-free = 4.61; F(2, 99) = 8.74, p < .01), and purchase intention (MPercentage-off = 4.26, MBuy-one-get-one-free = 3.34; F(2, 99) = 6.78, p < .05) compared to when offered a buy-one-get-one-free promotion, supporting H2b (see Fig. 1).
Fig. 1
Interaction effects of math anxiety and promotion framing types (Study 1)
×
4 Study 2
4.1 Method
To examine H3, a 2 (math anxiety levels: low vs. high) × 2 (discount frame formats: a single discount vs. multiple discounts) between-subjects factorial experiment was conducted with US consumers recruited from Amazon Mechanical Turk (MTurk) (n = 236, 43% female). 59% of these participants were aged 18–34, 30% were aged 35–49, and 11% aged 50 or older. Cell sizes ranged from 33 to 85. Each respondent received a $1 reward for participation.
The experimental product was a laptop computer based on pretest results. Participants were presented with a buying scenario involving purchasing a laptop from an online seller offering a deal. They were randomly assigned to either a single discount ($479.99 with 25% off) or multiple discounts ($479.99 with 10% off, an extra $32 off if purchased today, and an additional 10% off only for valued customers like you). In both conditions, the net price after discount was $359.99 (see Table 1). Participants then responded to the dependent measures identical to those employed in Study 1, with the level of math anxiety assessed using the MARS-Brief scale (see Table 3 in Appendix).
4.2 Results
The sample was divided into high and low math-anxiety groups using the mean math anxiety scores (NHigh = 155, NLow = 81). The high and low math-anxiety groups had significantly different levels of math anxiety (MHigh = 3.99, MLow = 2.20; F = 527.59, p < .001). There was a significant difference in the accuracy of final price calculations between the two discount frames (MSingle Discount = .40, MMultiple Discount = .19; F = 12.20, p < .01). In the single discount condition, 42% of low math-anxiety participants and 40% of high math-anxiety participants accurately calculated the final price. However, in the multiple discount condition, only 29% of low math-anxiety participants and 11% high math-anxiety participants provided correct answers (see Table 2).
A 2 (math anxiety levels: low vs. high) × 2 (discount frame formats: a single discount vs. multiple discounts) MANCOVA was conducted to test the interaction effect in H3. As a covariate, gender did not reveal a significant influence (F(4, 231) = .58, p > .05). The MANOVA showed a significant interaction effect (λ = .96; F(4, 231) = 2.60, p < .05). Subsequent ANOVA exhibited a significant interaction effect on perceived savings (F(4, 231) = 4.32, p < .05), price acceptability (F(4, 231) = 5.44, p < .05), and purchase intention (F(4, 231) = 10.49, p < .01).
When the discount promotion was framed as a single discount rather than multiple discounts, participants with high math-anxiety reported significantly higher perceived savings (MSingleDiscount = 6.02, MMultipleDiscount = 5.59; F(2, 152) = 10.62, p < .01), price acceptability (MSingleDiscount = 6.07, MMultipleDiscount = 5.75; F(2, 152) = 6.40, p < .05), and purchase intention (MSingleDiscount = 6.00, MMultipleDiscount = 5.66; F(2, 152) = 6.59, p <. 05), supporting H3a. Conversely, when multiple discounts were offered, as opposed to a single discount, participants with low math-anxiety showed no significant difference in perceived savings (MMultipleDiscount = 5.11, MSingleDiscount = 4.97; F(2, 78) = .24, p > .05), price acceptability (MMultipleDiscount = 5.41, MSingleDiscount = 5.11; F(2, 78) = 1.17, p > .05), and purchase intention (MMultipleDiscount = 5.07, MSingleDiscount = 4.39; F(2, 78) = 3.77, p > .05), supporting H3b (see Fig. 2).
Fig. 2
Interaction effects of math anxiety and discount framing formats (Study 2)
×
5 General discussion
Across two studies, we demonstrate that math anxiety impacts consumers’ perceived savings, price acceptability, and purchase intention, depending on how promotions or discounts are framed. While Study 1 shows that 44% of college students exhibit high levels of math anxiety, this proportion increases to 65% in Study 2 among general adult consumers. This may reveal a greater likelihood of math anxiety among older individuals and those not actively engaged in academic pursuits.
Study 1 examines the mediating effect of negative affect and shows that lower math ability indirectly heightens math anxiety by boosting negative affect. Consumers with lower math abilities struggle with price calculations, leading to negative emotions and increased anxiety. Furthermore, consumers with high math anxiety are less accurate in calculating final prices after promotions compared to individuals with low math anxiety, supporting the argument that negative emotions may weaken consumers’ ability to solve numerical problems (Reyes, 2019). The study also reveals that consumers with high math-anxiety perceive greater savings, price acceptability, and purchase intentions from buy-one-get-one-free promotion frames, as they do not require complex calculations. Conversely, those with low math-anxiety favor percentage-off bargains. When math-anxious consumers encounter a buy-one-get-one-free deal, they focus on the free products, which elicits psychological pleasure. They may choose a deal offering a free quantity, even if it is more costly (Nicolau, 2012). According to the attentional control theory, highly math-anxious individuals focus on a stimulus-driven system directed by emotions (Eysenck et al., 2007). Our results also corroborate findings that consumers’ preference for buy-one-get-one-free promotions over rebates is related to a reduced need for price computations (Davis & Millner, 2005).
Study 2 shows that consumers with high math-anxiety prefer price promotions with a single discount due to easier numerical processing. Multiple discounts, requiring complicated price computations to determine the final net price, were less appealing to them. Consumers with high math-anxiety may avoid such offers rather than devoting time and effort to calculate the overall savings (Chen & Rao, 2007). In contrast, the number of discounts does not significantly influence consumers with low math-anxiety; rather, research suggests that they would be more willing to purchase products with multiple discounts since they may perceive greater chances for savings and a higher transaction value (Grewal et al., 1998).
This research contributes to the pricing and marketing literature by understanding the interplay of math anxiety and promotion framing on consumer purchase decisions, particularly regarding perceived savings and price acceptability. While cognitive psychology suggests that math anxiety may reduce math ability in children and cause them to be reluctant to take math classes (Ashcraft & Moore, 2009), we show that math ability is an antecedent of math anxiety among adult consumers, especially when insufficient math skills lead to negative affect during shopping. Our findings provide empirical evidence for the literature that math anxiety reduces working memory capacity and individuals’ motivation to perform math calculations (Ashcraft & Kirk, 2001; Reyes, 2019).
Our results also align with previous research indicating that greater perceived savings and price acceptance correlate with higher purchase likelihoods (Xia & Monroe, 2004). Math anxiety adversely influences perceived savings and purchase intent, particularly with promotion frames requiring complex calculations to ascertain the final selling price or amount of savings. The interaction effects (H2 and H3) uncovered differences between consumers with high and low math anxiety in their responses to various promotion frames.
6 Implications
Pricing managers should exercise caution when using price presentations with low processing fluency, particularly for consumer segments with difficulty processing numerical information. Our Study 1 shows that offering free products for the same price is more appealing to people who dislike or have difficulty processing numerical information. The free product promotion framing might be beneficial for frequently purchased products that require minimum consumer involvement, such as laundry detergents and ordinary grocery products. As the elaboration likelihood model (ELM) indicates, in low involvement buying situations, consumers are less motivated to process information and more likely to depend on heuristics (Petty & Cacioppo, 1986). Low product price levels and perceived risk in low-involvement buying may dissuade math-anxious consumers from processing complex price computation, making offers such as buy-one-get-the-next-one-50%-off or similar promotions, where the price is not rounded or complex calculations are required, problematic (Coulter & Roggeveen, 2014).
Our Study 2 findings suggest that marketers could use a single discount promotion for products that highly math-anxious consumers often purchase, such as convenience goods, groceries, household products, and low-priced shopping goods like shoes and apparel. A single discount is easier to comprehend and induces higher perceived savings and price acceptability. However, for high-involvement purchases, consumers will have stronger motivation to process information and devote more effort to computing the net price (Suri et al., 2013). Therefore, it may be advantageous to sell electronics, personal computers, and other high-priced items to professionals or business consumers by offering multiple discounts that induce extra savings perceptions. If sellers wish to offer multiple discounts, we suggest that they use money-off rather than percentage-off discounts to simplify price computation.
Finally, we recommend sellers apply simple price promotion frames and employ tactics to reduce negative affect and increase perceived savings. Improving consumers’ mood and displaying the actual amount saved are two effective tactics. For example, researchers suggest that using slow-tempo background music helps alleviate math anxiety at the time of purchase. Furthermore, stressing the amount of savings, such as “you save $25.00,” might reduce the need to use arithmetic and boost the likelihood of making a sale (Feng et al., 2014).
7 Limitations and future research
The present study possesses several limitations that offer prospects for future research. First, our research examines a limited number of promotion frames and their interactions with math anxiety in influencing consumers’ price perceptions and purchase decisions. Future research may extend the findings to additional forms of price presentations. For example, math-anxious consumers may respond differently to odd-ending prices compared to even-ending or rounded prices. Second, we used two products and a set of hypothetical purchase scenarios in our two studies to explore participants’ price perceptions and responses. Future research could explore other products and scenarios to enhance understanding. Finally, the present study referred to an online consumer sample from MTurk to broaden the research participants beyond a student demographic. Future research may expand these studies to include in-store or shopping mall samples and real consumers to corroborate the results and identify any potential variations.
Declarations
Ethical approval
All procedures performed in the studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.
Consent to participate
Informed consent was obtained from all individual participants in the studies.
Conflict of interest
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.