Eco-Sensitive Minds: Clustering Readiness to Change and Environmental Sensitivity for Sustainable Engagement

Abstract

To counter the consequences of climate change on both planetary and human health, a greater adoption of sustainable behaviors is required. In this context, two factors emerge as potentially crucial: Readiness to Change (RTC) and environmental sensitivity. The study aimed to investigate the interaction between these two constructs and their impact on the engagement of pro-environmental behaviors and levels of eco-anxiety, in order to assess potential differences in behavioral and affective factors that may support the improvement of sustainable habits. Data were anonymously collected online from 947 participants. A Random Forest Clustering (RFC) analysis was performed as well as Analysis of Variance (ANOVA) to explore differences between the identified clusters in terms of sustainable behaviors and eco-anxiety. The RFC revealed the presence of seven distinct clusters and highlighted that environmental sensitivity plays a key role in defining them. Moreover, the findings showed that high RTC combined with high environmental sensitivity is associated with greater engagement in pro-environmental behaviors and higher levels of eco-anxiety. These results represent a promising groundwork for the development of both future studies in this field of research and targeted educational and awareness programs addressing the climate crisis.

1. Introduction

1.1. Research Background

1.1.1. Climate Change and Human Health: A Self-Perpetuating Cycle of Risk

Climate change consequences represent a challenge for both planetary and public health [1,2]. As highlighted by the COP28, global average temperatures are rising, leading to critical concerns related to the emergence of harmful climate risks on both ecosystems and humanity [3].

In this regard, previous scholars pointed out how human behavior plays a significant role on climate change and its associated issues—from greenhouse gas emissions to the adoption of unsustainable lifestyles (e.g., waste management) [4,5,6,7,8,9].

On the other hand, environmental consequences also have a negative impact on human health, both physical and mental [2,10,11,12,13,14]. In keeping with this, from a physical point of view, links between climate change and kidney, respiratory, and cardiovascular diseases have been observed [10,13].

Regarding mental health, the literature highlighted how climate risks and air pollution are potentially associated with conditions such as cognitive deficits or post-traumatic stress disorder [14]. In this scenario, among more strictly psychological symptoms, eco-anxiety has also been identified [15].

1.1.2. The Dual Role of Eco-Anxiety: Psychological Distress and Pro-Environmental Engagement

Despite several definitions of eco-anxiety (EA) in the literature, this kind of construct may be conceptualized as a persistent fear of environmental disaster as well as the perception of a deterioration of ecological systems directly connected to climate change [15,16,17,18]. Moreover, EA is increasingly interpreted as a multidimensional construct composed of different dimensions such as affective symptoms, rumination, behavioral symptoms, and anxiety about personal impact [19,20,21].

In this scenario, findings from the literature also pointed out the possibility that eco-anxiety may represent a proactive response to climate change considering its putative links with both higher engagement in sustainable behaviors and higher environmental sensitivity [22,23,24,25,26,27]. Notably, sensitive people exhibit higher emotional reactivity as well as higher empathy and compassion toward the natural environment and its preservation [28,29,30]. Accordingly, these characteristics may potentially be associated with a greater tendency to engage in sustainable behaviors and exacerbate a higher susceptibility to eco-anxiety symptoms [27,29,31,32,33].

1.1.3. The Central Role of Pro-Environmental Behaviors

In light of all the above, considering climate change consequences, the importance of increasing Pro-Environmental Behaviors (PEBs) becomes evident, as a way to address both environmental health issues and individual health. Notably, PEBs refer to the engagement of sustainable and useful actions to reduce and improve environmental harms (e.g., using energy-saving light bulbs) and well-being (e.g., by planting trees), respectively [34]. As suggested by Masud and colleagues [35], individual PEBs can include actions such as recycling extensively, opting for eco-friendly products, and minimizing waste generation. Additional examples are represented by turning off unnecessary lights, unplugging electronic devices when not in use, and replacing outdated appliances with energy-efficient alternatives. People may also engage in reusing containers and bottles, installing renewable energy systems in terms of solar panels or wind turbines, and using energy-saving light bulbs. In terms of mobility, behaviors include walking or cycling instead of driving, choosing public transportation over personal vehicles, reducing overall car usage, switching to more fuel-efficient cars, or buying vehicles with smaller engines.

Over the years, to promote engagement in pro-environmental behaviors, several theoretical frameworks have been developed to point out the potential psychological determinants and antecedents that may support and encourage the adoption of sustainable behaviors [36,37,38,39,40,41,42,43,44]. Among these, Readiness to Change (RTC) represents a relevant psychological construct [44,45].

1.1.4. Readiness to Change: A Conceptual Overview Within Environmental Applications

Over recent decades, the concept of Readiness to Change (RTC) has evolved from a peripheral element in behavior change theories to a central construct, particularly relevant in both clinical and sustainability domains. Classical behavior change models are typically categorized as either continuum or stage-based [46]. Continuum models, such as the Theory of Planned Behavior [47,48] and Protection Motivation Theory [49], emphasize intention as the principal predictor of behavior, yet have been criticized for their linear assumptions and limited consideration of post-intentional processes [46,50]. Stage models, particularly the Transtheoretical Model [36,37], propose discrete behavioral phases (e.g., precontemplation, contemplation, preparation, action, maintenance), but these, too, have been contested for their arbitrary divisions and limited predictive power [40,51,52]. Therefore, RTC can be conceptualized either as a state, reflecting the current level of readiness of an individual [53], or as a process that unfolds over time [54]; notably, Dalton and Gottlieb [55] proposed an integrative perspective that combines both views, acknowledging the dynamic nature of readiness. Due to the aforementioned limitations, scholars pointed out the need for more integrative frameworks. In keeping with this, RTC represents a multifaceted construct encompassing cognitive, affective, and behavioral readiness [55,56,57,58] and models like HAPA [46] and SOCRATES [59] emphasize constructs such as risk perception, outcome expectancy, and self-efficacy. In line with this, Duradoni and colleagues [44] proposed a new conceptualization for RTC in order to capture its multidimensional nature by moving beyond the unidimensional perspectives. This novel framework was specifically tailored to the domain of PEBs, an area in which the application of the RTC construct remains notably scarce and fragmented. Previous studies have typically addressed RTC in environmental contexts using inconsistent definitions and without integrating the range of psychological components that may be involved. In response to these limitations, the model developed by Duradoni et al. [44] offers a structured and comprehensive description of the multiple psychological factors that interact in shaping sustainable behavioral change. The model identifies seven key dimensions characterized by empirical relevance in the domain of sustainability:

(a)

Perceived importance of the problem: particularly in the form of risk perception, is a strong driver of environmental concern. Individuals who perceive climate crisis as important are more likely to engage in mitigation behaviors [60,61].

(b)

Motivation: especially when intrinsic or self-determined, plays a critical role in promoting voluntary, consistent PEB [62,63].

(c)

Self-efficacy: the belief in one’s capacity to implement change, is consistently linked with stronger pro-environmental intentions and actions [61,64].

(d)

Effectiveness of the proposed solution: conceptualized as response efficacy, refers to the belief that one’s actions can meaningfully reduce environmental harm and it is an essential predictor of adaptive behavior [65].

(e)

Social support: understood through constructs such as individual and collective social capital, reinforces environmental action by shaping norms and enhancing perceived collective efficacy [66,67].

(f)

Action and involvement: are supported by the spillover effect—prior engagement in green behaviors increases the likelihood of further action by reinforcing pro-environmental self-identity [68,69].

(g)

Perceived readiness: the subjective sense of being mentally and practically prepared to act sustainably has been shown to influence behaviors like green purchasing, especially when grounded in internal drivers such as self-identity and responsibility [70,71].

Particularly, the multidimensional approach above allowed for a more nuanced understanding of the change process, emphasizing the complex interplay between cognitive, motivational, and contextual variables in the promotion of PEBs. However, despite the importance of the seven factors above, the literature investigating RTC in the context of sustainability remains limited and fragmented. In keeping with this, the multidimensional framework by Duradoni et al. [44], grounded in both classical theory and empirical evidence, represents a promising effort to consolidate this field of research and by offering a strong structure for assessing individual readiness across different sustainable contexts. In particular, the results from the existing literature have highlighted the ability of RTC to assess both individual and collective commitment to sustainable behaviors, providing valuable data and information for the implementation of effective, sustainable intervention programs [44]. In particular, unlike the other valid theoretical models mentioned above, the ability to also measure collective commitment to combating climate change makes it an advantageous model that combines individual and group-level perspectives [44].

1.2. Literature Review and Research Gap: Readiness to Change and Sensitivity

As previously mentioned, sensitivity appears to be associated with greater involvement in pro-environmental behaviors as well as with higher levels of eco-anxiety [27]. Although several studies have highlighted a correlation between the sensitivity trait and the Behavioral Inhibition System (BIS) [72,73,74], authors have reported contrasting evidence, linking this trait instead to the Behavioral Activation System (BAS) [73,75,76]. Notably, Smolewska and colleagues [76] observed a correlation between sensitivity trait and reward responsiveness, by reflecting the intensity with which rewards elicit positive emotional responses. Based on this, it is plausible to assume that highly sensitive people are more inclined to feel positive emotions in response to rewarding stimuli. Considering the link between sensitivity and connectedness to nature, such rewarding stimuli may also include actions and behaviors for environmental preservation [27,76,77]. This interpretation could further support the link between high sensitivity and PEBs implementation [27]. Moreover, taking into consideration the association between high sensitivity and sustainable attitudes [27], and in line with the Theory of Reasoned Action (TRA) [78], it is possible to speculate about the possibility that such attitudes may also promote the development of behavioral intentions [27]. Furthermore, by strictly referring to the sustainable context, environmental sensitivity is linked with environmental citizenship behavior—a dimension that RTC seems to be able to measure [44,79]. In keeping with this, from our perspective, it is possible to assume a relationship between environmental sensitivity and RTC. Moreover, the above relationship is also supported by other results in the literature, considering that RTC appears to be also capable of measuring individual aspects as well [44]. In fact, several scholars have observed that environmental sensitivity can also have positive effects on pro-environmental attitudes (e.g., buying sustainable products) [80,81], which can theoretically derive from the dimensions of RTC. Moreover, as previously mentioned, from a more strictly behavioral perspective, by considering the links between sensitivity and Readiness to Change and pro-environmental attitudes [27,44], it can be assumed that the combination of the two investigated phenomena may strengthen the involvement in sustainable behaviors. Based on this perspective, sensitivity combined with a readiness to change may also have an impact on eco-anxiety, considering that both constructs appear to be associated with this phenomenon [27,82]. Specifically, a high sensitivity associated with a strong readiness to change can favor sustainable engagement from both behavioral and emotional perspectives. In summary, considering the potential behavioral predisposition and intentions of highly sensitive individuals, it can be assumed that this kind of trait may be related to the construct of RTC in terms of a psychological dimension that reflects the willingness of people to engage in behavioral change [44]. In keeping with this, Duradoni and colleagues [27] pointed out the need for further research to examine whether RTC dimensions are putatively linked with environmental sensitivity traits [27,44].

1.3. Aims of the Study: An Innovative Perspective

In light of all the above, the present work aimed to explore how high sensitivity clusters in relation to Readiness to Change. Moreover, the present work also aimed at investigating differences between clusters concerning the implementation of pro-environmental behaviors and eco-anxiety levels. These latter two constructs were selected as outcome factors, as they represent, respectively, a behavioral dimension and a more affective one, both of which are essential in efforts to counteract the consequences of climate change by postulating the following hypothesis [32,83,84,85]:

H1. 

The combination of high levels of sensitivity and RTC will promote greater engagement in sustainable behaviors as well as in eco-anxiety.

Notably, to the best of our knowledge, considering the lack in the literature concerning the potential links between sensitivity and RTC as well as the capability of the combination of these two constructs in strengthening the engagement in sustainable behaviors and eco-anxiety, the present work takes a step forward to the comprehension of these kinds of phenomena by providing the groundwork for the implementation of future studies in this field of research.

2. Materials and Methods

2.1. Data Collection

Data were collected between April 2023 and May 2024 through an online survey promoted on the main-known social networks which anonymity promoted a lower risk of social desirability bias [86]. Data were collected in accordance with the privacy laws of Italy (Law Decree DL-101/2018) and the EU regulations (2016/679). The final sample was composed of 947 people (31.9% cis-gender males; 63.4% cis-gender females; 4.7% people belonging to the LGBTQIA + community) with mean age equal to 25.5 years (sd = 9.12). In the majority of cases, participants have a Bachelor’s Degree (57.7%), are students (72.2%), and a reported income between €10,000 and €40,000 (59%) (

Table 1. Socio-demographic characteristics of the sample.

	Percentage
Educational level
Primary school certificate	0.3
Middle school certificate	2.3
High school diploma	57.7
Bachelor’s Degree	29.8
Master’s Degree	7.2
University Master’s Program	1.0
Postgraduate specialization/PhD	1.7
Employment Status
Employed	24.3
Unemployed	2.6
Student	72.2
Retired	0.8
Income
Less than €10,000	1.0
Between €10,000 and €40,000	59.0
Between €40,000 and €70,000	31.4
Between €70,000 and €120,000	6.4

).

2.2. Instruments

In accordance with the aim of the study, the following measures were collected (

Table 2. Synthesis of the instruments used.

Instrument	Author(s)	Aim
Highly Sensitive Person Scale	Aron and Aron [29] Smolewska et al. [76]	Assessing sensitivity traits in terms of: (a) Ease of Excitation, (b) Aesthetic Sensitivity, and (c) Low Sensory Threshold.
The Readiness to Change Scale	Duradoni et al. [44]	Measuring dimensions of Readiness to Change, namely (a) perceived importance of the problem, (b) motivation, (c) self-efficacy, (d) effectiveness of the proposed solution, (e) social support, (f) action, and (g) perceived readiness.
Hogg Eco-Anxiety Scale	Hogg et al. [20]	Assessing eco-anxiety in terms of (a) affective symptoms, (b) rumination, (c) behavioral symptoms, and (d) anxiety for personal impact.
The Pro-environmental Behavior Scale	Markle [87]	Measuring pro-environmental behaviors such as (a) conservation, (b) environmental citizenship, (c) food, and (d) transportation.

):

Highly Sensitive Person Scale (HSPS) [29,76]: This scale consists of 27 items designed to assess the sensitivity trait, rated on a 7-point Likert scale (1 = “Not at all” to 7 = “Completely”). In the present study, the 18-item version of the scale of Smolewska and colleagues [76] was used. The instrument measures both an overall sensitivity score and three specific dimensions: (1) Ease of Excitation (EOE), (2) Aesthetic Sensitivity (AES), and (3) Low Sensory Threshold (LST) [76]. Overall, the scale demonstrated good internal consistency, with a Cronbach’s alpha of 0.89 (EOE = 0.81; AES = 0.72; LST = 0.78).

The Readiness to Change Scale [44]: The scale assess seven different dimensions of subjective RTC in terms of perceived importance of the problem (RTC-PI), motivation (RTC-M), self-efficacy (RTC-SE), effectiveness of the proposed solution (RTC-ES), social support (RTC-SS), action (RTC-A), and perceived readiness (RTC-PR) [44]. All the dimensions have a good internal consistency ranging from 0.81 to 0.87 (McDonald’s ω) [44]. The instrument is characterized by 29 items scored on a 5-point Likert scale (1: “strongly disagree”; 5: “strongly agree”) [44].

Hogg Eco-Anxiety Scale (HEAS) [20]: The instrument assess level of eco-anxiety through 13 items scored on a 4-point Likert scale (0: “Not at all”; 3: “Nearly every day”) [20]. The scale measures four dimensions in terms of affective symptoms (AS), rumination (R), behavioral symptoms (BS), and anxiety for personal impact (API) [20]. The instrument has a very good internal consistency (Cronbach’s alpha = 0.93) [20].

The Pro-environmental Behavior Scale (PEB) [87]: The scale is composed of 19 items able to assess multiple pro-environmental behavior in terms of conservation, environmental citizenship, food habits, and transportation choices. The items are rated using various Likert-type formats. Specifically, items 1 through 6 are measured on a 5-point scale (1 = “never”, 5 = “always”); item 7 uses a 3-point scale (1 = “very high”, 3 = “low”); items 8, 9, and 12 are scored dichotomously (1 = “yes”, 5 = “no”); items 10 and 11 also use a 5-point scale (1 = “never”, 5 = “constantly”); item 13 is evaluated on a 5-point scale based on numerical categories (1 = “24 or less”, 5 = “40 or more”); items 14 to 16 are coded dichotomously (1 = “no”, 5 = “yes”); and items 17 to 19 are rated on a 3-point scale (1 = “never”, 5 = “frequently”). Duradoni and colleagues [44] adapted the Italian version of the scale demonstrating satisfactory internal consistency (Cronbach’s α = 0.76).

2.3. Data Analysis

In order to automatically generate clusters incorporating factors such as readiness to change and environmental sensitivity, a Random Forest Clustering (RFC) procedure was conducted [88,89,90]. Specifically, the three dimensions of the HSPS and the seven dimensions of the RTC scale were used as input variables. The proximity matrix was used as clustering technique. The model was optimized on the basis of the Bayesian Information Criterion (BIC) and, based on the seven RTC dimensions, the maximum node number was fixed at seven [44]. Specifically, both the BIC and R² values were used to assess the goodness-of-fit of the model [91,92]. Moreover, the Gini index was extracted to determine which factors mostly contributed to the differentiation between clusters [88]. Due its methodological characteristics, the use of RFC had numerous advantages such as the ability to capture complex non-linear relationships, its strong performance in the presence of high-dimensional data, and its robustness with small sample sizes [89,90,93,94]. Furthermore, the RFC outputs were subsequently used as an independent variable in Analysis of Variance (ANOVA) to examine differences among clusters in PEBs engagement and eco-anxiety values, using a significance threshold of p < 0.05. Moreover, a post hoc analysis (Tukey’s test). The proportion of variance explained by the independent variable was quantified using Eta-squared (η²), with thresholds defined as follows: small (η² ≤ 0.01), medium (η² ≈ 0.06), and large (η² ≥ 0.14) [95]. Finally, a multiple stepwise linear regression analysis assuming pro-environmental behaviors and eco-anxiety as dependent variables and RTC and sensitivity dimensions as independent variables was performed. Notably, different models for each pro-environmental behaviors and eco-anxiety factors were carried out in order to observed which RTC and sensitivity dimensions better predicted the phenomena above. Statistical analyses were performed using JASP software (version 0.19.0.0) and Statistical Package for the Social Sciences (SPSS) software (version 23).

3. Results

3.1. Random Clustering Model: Results

The RFC pointed out a total of seven clusters with R² (proportion of explained variance) and BIC equal to 39.6% and 6189.900, respectively (Figure 1). Concerning cluster characteristics (

Table 3. Cluster information and model performance metrics.

Cluster	1	2	3	4	5	6	7
Size	455	148	141	59	61	28	54
Explained proportion within-cluster heterogeneity	0.466	0.129	0.183	0.087	0.075	0.027	0.033
Within sum of squares	2658.355	734.819	1046.305	498.609	427.623	156.929	187.602
Pearson’s γ	0.371
Dunn index	0.078
Entropy	1.543
Calinski-Harabasz index	102.495

), the results showed a low Dunn Index (0.078) [96]; however, the performance metrics highlighted a high Calinski–Harabasz Index (102.495), indicating the presence of a well-defined segmentation [97]. Moreover, in

Table 4. Cluster means.

	RTC-PI	RTC-M	RTC-SE	RTC-ES	RTC-SS	RTC-A	RTC-PR	HSPS-EOE	HSPS-AES	HSPS-LST
Cluster 1	0.146	0.190	0.076	0.015	0.091	0.126	0.277	−0.054	−0.050	−0.185
Cluster 2	0.424	0.428	0.318	0.478	0.158	0.543	0.423	0.930	0.977	1.000
Cluster 3	−0.636	−0.872	−0.616	−0.426	−0.386	−0.641	−0.984	−0.248	−0.408	−0.148
Cluster 4	−1.469	−1.506	−1.425	−1.329	−0.981	−1.609	−1.585	−0.856	−1.017	−0.660
Cluster 5	1.419	1.630	1.527	1.521	0.901	1.408	1.349	0.364	0.726	0.434
Cluster 6	−0.833	−1.077	−1.255	−1.332	−1.032	−1.250	−1.610	1.222	0.637	1.027
Cluster 7	−0.295	−0.136	0.582	0.099	0.400	−0.062	0.119	−1.554	−1.226	−1.094

Notes: RTC, Readiness to Change; PI, Perceived importance, M, Motivation, SE, Self-efficacy; ES, Effectiveness of the proposed solution; SS, Social support; A, Actions; PR, Perceived readiness; HSPS, Highly Sensitive Person Scale; EOE, Ease of Excitation; AES, Aesthetic Sensitivity; LST, Low Sensory Threshold.

and Figure 2, values and standard deviations values of the investigated factors per cluster. In detail, Cluster 1 is characterized by positive values of RTC and negative values of sensitivity. On the other hand, Clusters 2 and 5 are composed of positive values in both investigated domains. On the contrary, Clusters 3 and 4 are characterized by negative values of both RTC and sensitivity, whereas Cluster 6 is composed of negative RTC values but positive sensitivity values. Finally, Cluster 7 is the most heterogeneous, considering that it has both positive and negative values in the RTC domain and negative values in the sensitivity domain. Finally, on the basis of the Gini Index (mean decrease in accuracy; MDA), the most important factor in distinguishing the clusters was HSPS-EOE (MDA:116.559) (

Table 5. Gini Index.

	Mean Decrease in Gini Index
HSPS-EOE	116.559
RTC-M	104.388
HSPS-LST	101.586
RTC-PI	95.845
RTC-PR	94.026
HSPS-AES	93.166
RTC-SE	91.284
RTC-A	90.774
RTC-ES	82.064
RTC-SS	74.996

Notes: RTC, Readiness to Change; PI, Perceived importance, M, Motivation, SE, Self-efficacy; ES, Effectiveness of the proposed solution; SS, Social support; A, Actions; PR, Perceived readiness; HSPS, Highly Sensitive Person Scale; EOE, Ease of Excitation; AES, Aesthetic Sensitivity; LST, Low Sensory Threshold.

).

3.2. ANOVA: Results

Concerning the ANOVA outputs, results pointed out significant differences with a medium proportion of variance explained between clusters for what concern the implementation of pro-environmental behaviors in terms of conservation (F = 16.900; η² = 0.099; p < 0.001), environmental citizenship (F = 23.438; η² = 0.137, p < 0.001), food (F = 16.649; η² = 0.097 p < 0.001), and transportation (F = 8.313; η² = 0.051; p < 0.001) (

Table 6. ANOVA results concerning pro-environmental behaviors (PEB).

	F	p	η2	95% CI for η2
				Lower	Upper
PEB-C
Clusters	16.900	<0.001	0.099	0.061	0.132
PEB-EC
Clusters	23.438	<0.001	0.137	0.094	0.175
PEB-F
Clusters	16.649	<0.001	0.097	0.060	0.129
PEB-T
Clusters	8.313	<0.001	0.051	0.023	0.076

Notes: PEB, Pro-environmental behaviors, C, Conservation; EC, Environmental citizenship; F, Food; T, Transportation.

; Figure 3).

Notably, Tukey’s post hoc correction revealed significant differences at p < 0.05 between investigated clusters concerning PEB-C (

Table A2. Post hoc comparison concerning PEB-Conservation.

			95% CI for Mean Difference					95% CI for Cohen’s d
		Mean Difference	Lower	Upper	SE	t	Cohen’s d	Lower	Upper	ptukey
1	2	−1.357	−2.347	−0.367	0.335	−4.049	−0.387	−0.679	−0.095	0.001
	3	1.392	0.386	2.398	0.341	4.089	0.397	0.100	0.694	<0.001
	4	2.195	0.759	3.630	0.486	4.518	0.626	0.201	1.050	<0.001
	5	−2.512	−3.937	−1.087	0.482	−5.210	−0.716	−1.138	−0.294	<0.001
	6	1.252	−0.767	3.271	0.683	1.833	0.357	−0.237	0.951	0.526
	7	0.306	−1.186	1.799	0.505	0.607	0.087	−0.352	0.526	0.997
2	3	2.749	1.519	3.979	0.416	6.603	0.784	0.418	1.150	<0.001
	4	3.552	1.951	5.152	0.542	6.557	1.013	0.537	1.488	<0.001
	5	−1.155	−2.746	0.436	0.538	−2.146	−0.329	−0.798	0.139	0.327
	6	2.609	0.470	4.748	0.724	3.604	0.744	0.113	1.375	0.006
	7	1.663	0.011	3.316	0.559	2.975	0.474	−0.013	0.961	0.047
3	4	0.802	−0.808	2.413	0.545	1.472	0.229	−0.245	0.702	0.761
	5	−3.904	−5.505	−2.303	0.542	−7.206	−1.113	−1.590	−0.636	<0.001
	6	−0.140	−2.287	2.007	0.727	−0.193	−0.040	−0.671	0.591	1.000
	7	−1.086	−2.748	0.576	0.562	−1.931	−0.310	−0.799	0.179	0.460
4	5	−4.707	−6.607	−2.807	0.643	−7.319	−1.342	−1.909	−0.775	<0.001
	6	−0.942	−3.321	1.436	0.805	−1.171	−0.269	−0.968	0.431	0.905
	7	−1.888	−3.840	0.064	0.661	−2.859	−0.538	−1.113	0.037	0.065
5	6	3.764	1.392	6.136	0.803	4.689	1.073	0.372	1.775	<0.001
	7	2.819	0.875	4.763	0.658	4.284	0.804	0.229	1.378	<0.001
6	7	−0.946	−3.359	1.468	0.817	−1.158	−0.270	−0.979	0.440	0.910

), PEB-EC (

Table A4. Post hoc comparison concerning PEB-Environmental citizenship.

			95% CI for Mean Difference					95% CI for Cohen’s d
		Mean Difference	Lower	Upper	SE	t	Cohen’s d	Lower	Upper	ptukey
1	2	−0.972	−2.026	0.082	0.357	−2.725	−0.267	−0.567	0.032	0.093
	3	1.341	0.281	2.401	0.359	3.738	0.369	0.067	0.671	0.004
	4	2.971	1.457	4.486	0.512	5.798	0.817	0.384	1.251	<0.001
	5	−3.617	−5.143	−2.091	0.517	−7.003	−0.995	−1.434	−0.556	<0.001
	6	2.726	0.557	4.896	0.734	3.713	0.750	0.132	1.368	0.004
	7	1.265	−0.313	2.842	0.534	2.369	0.348	−0.100	0.796	0.213
2	3	2.313	1.010	3.616	0.441	5.246	0.636	0.264	1.008	<0.001
	4	3.943	2.250	5.637	0.573	6.881	1.085	0.598	1.571	<0.001
	5	−2.645	−4.349	−0.941	0.577	−4.587	−0.728	−1.214	−0.241	<0.001
	6	3.698	1.400	5.996	0.778	4.755	1.017	0.361	1.673	<0.001
	7	2.237	0.487	3.987	0.592	3.777	0.615	0.117	1.113	0.003
3	4	1.630	−0.067	3.327	0.574	2.839	0.448	−0.034	0.931	0.069
	5	−4.958	−6.666	−3.250	0.578	−8.580	−1.364	−1.858	−0.869	<0.001
	6	1.385	−0.916	3.686	0.779	1.779	0.381	−0.272	1.034	0.562
	7	−0.076	−1.830	1.677	0.593	−0.128	−0.021	−0.518	0.476	1.000
4	5	−6.588	−8.610	−4.567	0.684	−9.631	−1.812	−2.400	−1.224	<0.001
	6	−0.245	−2.788	2.298	0.860	−0.285	−0.067	−0.788	0.654	1.000
	7	−1.706	−3.767	0.354	0.697	−2.448	−0.469	−1.055	0.116	0.180
5	6	6.343	3.794	8.893	0.863	7.352	1.745	1.011	2.479	<0.001
	7	4.882	2.813	6.951	0.700	6.973	1.343	0.748	1.938	<0.001
6	7	−1.462	−4.042	1.119	0.873	−1.674	−0.402	−1.134	0.330	0.634

), PEB-F (

Table A6. Post hoc comparison concerning PEB-Food.

			95% CI for Mean Difference					95% CI for Cohen’s d
		Mean Difference	Lower	Upper	SE	t	Cohen’s d	Lower	Upper	ptukey
1	2	−1.833	−3.171	−0.496	0.453	−4.051	−0.387	−0.679	−0.095	0.001
	3	1.266	−0.086	2.618	0.457	2.768	0.267	−0.027	0.561	0.083
	4	3.996	2.057	5.936	0.656	6.088	0.843	0.417	1.269	<0.001
	5	−2.822	−4.747	−0.896	0.652	−4.331	−0.595	−1.016	−0.174	<0.001
	6	1.340	−1.389	4.069	0.924	1.451	0.283	−0.311	0.876	0.774
	7	1.134	−0.884	3.151	0.683	1.660	0.239	−0.200	0.678	0.643
2	3	3.099	1.442	4.757	0.561	5.526	0.654	0.290	1.017	<0.001
	4	5.830	3.666	7.993	0.732	7.960	1.229	0.751	1.708	<0.001
	5	−0.989	−3.140	1.162	0.728	−1.358	−0.208	−0.676	0.259	0.824
	6	3.173	0.281	6.066	0.979	3.242	0.669	0.039	1.300	0.021
	7	2.967	0.733	5.201	0.756	3.924	0.626	0.138	1.113	0.002
3	4	2.730	0.557	4.903	0.735	3.713	0.576	0.102	1.050	0.004
	5	−4.088	−6.248	−1.928	0.731	−5.592	−0.862	−1.336	−0.389	<0.001
	6	0.074	−2.825	2.973	0.981	0.075	0.016	−0.615	0.646	1.000
	7	−0.132	−2.375	2.110	0.759	−0.174	−0.028	−0.515	0.460	1.000
4	5	−6.818	−9.387	−4.249	0.869	−7.841	−1.438	−2.005	−0.870	<0.001
	6	−2.656	−5.872	0.560	1.088	−2.441	−0.560	−1.260	0.140	0.183
	7	−2.863	−5.502	−0.224	0.893	−3.205	−0.604	−1.179	−0.028	0.024
5	6	4.162	0.955	7.369	1.085	3.835	0.878	0.178	1.578	0.003
	7	3.956	1.327	6.584	0.890	4.447	0.834	0.260	1.409	<0.001
6	7	−0.206	−3.470	3.057	1.104	−0.187	−0.044	−0.753	0.666	1.000

), and PEB-T (

Table A8. Post hoc comparison concerning PEB-Transportation.

			95% CI for Mean Difference					95% CI for Cohen’s d
		Mean Difference	Lower	Upper	SE	t	Cohen’s d	Lower	Upper	ptukey
1	2	−0.751	−1.514	0.013	0.258	−2.905	−0.278	−0.571	0.014	0.057
	3	0.659	−0.111	1.428	0.260	2.530	0.244	−0.050	0.539	0.150
	4	1.590	0.486	2.694	0.374	4.255	0.589	0.165	1.013	<0.001
	5	−0.287	−1.382	0.809	0.371	−0.773	−0.106	−0.525	0.313	0.987
	6	0.442	−1.111	1.995	0.526	0.841	0.164	−0.430	0.757	0.981
	7	1.236	0.087	2.384	0.389	3.180	0.458	0.018	0.898	0.025
2	3	1.409	0.465	2.354	0.320	4.409	0.522	0.159	0.885	<0.001
	4	2.340	1.108	3.573	0.417	5.611	0.867	0.392	1.342	<0.001
	5	0.464	−0.761	1.689	0.415	1.119	0.172	−0.296	0.640	0.922
	6	1.192	−0.454	2.839	0.557	2.140	0.442	−0.188	1.072	0.330
	7	1.986	0.714	3.258	0.431	4.613	0.736	0.247	1.225	<0.001
3	4	0.931	−0.305	2.167	0.418	2.225	0.345	−0.128	0.818	0.283
	5	−0.945	−2.174	0.284	0.416	−2.273	−0.350	−0.821	0.120	0.258
	6	−0.217	−1.866	1.433	0.558	−0.388	−0.080	−0.711	0.550	1.000
	7	0.577	−0.699	1.853	0.432	1.336	0.214	−0.274	0.702	0.835
4	5	−1.876	−3.338	−0.415	0.495	−3.793	−0.695	−1.256	−0.135	0.003
	6	−1.148	−2.977	0.682	0.619	−1.854	−0.425	−1.125	0.274	0.512
	7	−0.354	−1.856	1.147	0.508	−0.697	−0.131	−0.705	0.443	0.993
5	6	0.729	−1.096	2.553	0.618	1.180	0.270	−0.427	0.968	0.902
	7	1.522	0.027	3.018	0.506	3.008	0.564	−0.009	1.137	0.043
6	7	0.794	−1.063	2.650	0.628	1.263	0.294	−0.416	1.004	0.869

). In particular, for all sustainable behaviors considered, the differences between Clusters with a significant differences at p < 0.001 were between Clusters 1 and 4, Clusters 2 and 3, and Clusters 2 and 4.

Regarding results about eco-anxiety, the ANOVA pointed out significant differences with a large and medium proportion of variance explained between clusters in terms of HEAS-AS (F = 16.397; η² = 0.096; p < 0.001), HEAS-R (F = 19.453; η² = 0.112, p < 0.001), HEAS-BS (F = 9.397; η² = 0.057; p < 0.001), and HEAS-API (F = 28.043; η² = 0.154; p < 0.001) (

Table 7. ANOVA results concerning eco-anxiety (HEAS).

	F	p	η2	95% CI for η2
				Lower	Upper
HEAS-AS
Clusters	16.397	<0.001	0.096	0.059	0.128
HEAS-R
Clusters	19.453	<0.001	0.112	0.073	0.146
HEAS-BS
Clusters	9.397	<0.001	0.057	0.027	0.084
HEAS-API
Clusters	28.043	<0.001	0.154	0.110	0.193

Notes: HEAS, Hogg Eco-Anxiety Scale; AS, Affective symptoms; R, Rumination; BS, Behavioral symptoms; API, Anxiety for personal impact.

; Figure 4). Moreover, Tukey’s post hoc correction pointed out significant differences at p < 0.05 between investigated clusters concerning HEAS-AS (

Table A10. Post hoc comparison concerning HEAS-Affective symptoms.

			95% CI for Mean Difference					95% CI for Cohen’s d
		Mean Difference	Lower	Upper	SE	t	Cohen’s d	Lower	Upper	ptukey
1	2	−1.387	−2.134	−0.640	0.253	−5.484	−0.521	−0.813	−0.229	<0.001
	3	0.225	−0.536	0.987	0.258	0.875	0.085	−0.210	0.380	0.976
	4	1.494	0.397	2.591	0.371	4.024	0.562	0.135	0.988	0.001
	5	−1.128	−2.201	−0.055	0.363	−3.107	−0.424	−0.841	−0.007	0.032
	6	−0.417	−1.949	1.114	0.518	−0.805	−0.157	−0.750	0.437	0.984
	7	1.919	0.786	3.051	0.383	5.006	0.721	0.279	1.163	<0.001
2	3	1.612	0.684	2.541	0.314	5.131	0.606	0.244	0.968	<0.001
	4	2.881	1.662	4.100	0.413	6.983	1.083	0.604	1.561	<0.001
	5	0.259	−0.939	1.456	0.405	0.638	0.097	−0.367	0.561	0.996
	6	0.969	−0.652	2.590	0.549	1.767	0.364	−0.264	0.993	0.571
	7	3.305	2.054	4.556	0.423	7.807	1.242	0.750	1.735	<0.001
3	4	1.268	0.041	2.496	0.415	3.053	0.477	−0.0001385	0.954	0.038
	5	−1.354	−2.560	−0.148	0.408	−3.316	−0.509	−0.978	−0.040	0.016
	6	−0.643	−2.270	0.985	0.551	−1.167	−0.242	−0.873	0.389	0.906
	7	1.693	0.434	2.952	0.426	3.973	0.636	0.146	1.126	0.001
4	5	−2.622	−4.064	−1.180	0.488	−5.374	−0.986	−1.549	−0.423	<0.001
	6	−1.911	−3.720	−0.102	0.612	−3.122	−0.718	−1.421	−0.016	0.030
	7	0.425	−1.062	1.911	0.503	0.844	0.160	−0.417	0.736	0.980
5	6	0.711	−1.084	2.505	0.607	1.170	0.267	−0.429	0.963	0.905
	7	3.047	1.578	4.516	0.497	6.129	1.145	0.570	1.720	<0.001
6	7	2.336	0.505	4.167	0.620	3.770	0.878	0.166	1.590	0.003

), HEAS-R (

Table A12. Post hoc comparison concerning HEAS-Rumination.

			95% CI for Mean Difference					95% CI for Cohen’s d
		Mean Difference	Lower	Upper	SE	t	Cohen’s d	Lower	Upper	ptukey
1	2	−0.732	−1.264	−0.201	0.180	−4.070	−0.387	−0.678	−0.096	<0.001
	3	0.445	−0.097	0.987	0.183	2.428	0.235	−0.060	0.531	0.188
	4	1.406	0.619	2.192	0.266	5.280	0.743	0.311	1.174	<0.001
	5	−1.114	−1.883	−0.345	0.260	−4.282	−0.589	−1.010	−0.168	<0.001
	6	1.002	−0.088	2.092	0.369	2.718	0.529	−0.065	1.124	0.095
	7	1.373	0.567	2.178	0.273	5.033	0.725	0.283	1.167	<0.001
2	3	1.178	0.517	1.838	0.224	5.267	0.622	0.260	0.984	<0.001
	4	2.138	1.265	3.011	0.295	7.238	1.129	0.647	1.611	<0.001
	5	−0.382	−1.239	0.475	0.290	−1.317	−0.202	−0.669	0.265	0.844
	6	1.735	0.581	2.888	0.390	4.444	0.916	0.285	1.548	<0.001
	7	2.105	1.215	2.995	0.301	6.987	1.112	0.621	1.603	<0.001
3	4	0.961	0.082	1.840	0.297	3.229	0.507	0.027	0.987	0.022
	5	−1.560	−2.423	−0.696	0.292	−5.338	−0.824	−1.297	−0.350	<0.001
	6	0.557	−0.601	1.715	0.392	1.422	0.294	−0.337	0.925	0.790
	7	0.928	0.031	1.824	0.303	3.058	0.490	0.0006589	0.979	0.037
4	5	−2.520	−3.555	−1.485	0.350	−7.197	−1.331	−1.902	−0.760	<0.001
	6	−0.404	−1.695	0.888	0.437	−0.924	−0.213	−0.916	0.490	0.969
	7	−0.033	−1.095	1.029	0.360	−0.092	−0.018	−0.596	0.561	1.000
5	6	2.117	0.836	3.397	0.433	4.885	1.118	0.416	1.820	<0.001
	7	2.487	1.438	3.536	0.355	7.003	1.314	0.735	1.893	<0.001
6	7	0.370	−0.932	1.673	0.441	0.840	0.196	−0.514	0.905	0.981

), HEAS-BS (

Table A14. Post hoc comparison concerning HEAS-Behavioral symptoms.

			95% CI for Mean Difference					95% CI for Cohen’s d
		Mean Difference	Lower	Upper	SE	t	Cohen’s d	Lower	Upper	ptukey
1	2	−0.857	−1.492	−0.223	0.215	−3.995	−0.380	−0.671	−0.089	0.001
	3	−0.015	−0.661	0.631	0.219	−0.069	−0.007	−0.302	0.288	1.000
	4	0.740	−0.191	1.671	0.315	2.348	0.328	−0.098	0.753	0.222
	5	−0.645	−1.555	0.266	0.308	−2.092	−0.286	−0.702	0.131	0.358
	6	0.103	−1.219	1.426	0.447	0.231	0.046	−0.558	0.649	1.000
	7	1.528	0.559	2.497	0.328	4.659	0.677	0.232	1.122	<0.001
2	3	0.842	0.054	1.630	0.267	3.158	0.373	0.012	0.734	0.027
	4	1.597	0.563	2.632	0.350	4.562	0.707	0.232	1.182	<0.001
	5	0.213	−0.804	1.229	0.344	0.618	0.094	−0.370	0.558	0.996
	6	0.961	−0.436	2.358	0.473	2.032	0.425	−0.213	1.064	0.395
	7	2.386	1.317	3.455	0.362	6.594	1.056	0.563	1.550	<0.001
3	4	0.755	−0.287	1.797	0.353	2.142	0.334	−0.142	0.811	0.329
	5	−0.630	−1.653	0.394	0.346	−1.817	−0.279	−0.747	0.189	0.537
	6	0.119	−1.284	1.521	0.475	0.250	0.052	−0.588	0.693	1.000
	7	1.543	0.467	2.619	0.364	4.238	0.684	0.190	1.177	<0.001
4	5	−1.385	−2.608	−0.161	0.414	−3.344	−0.613	−1.174	−0.053	0.015
	6	−0.637	−2.191	0.918	0.526	−1.210	−0.282	−0.992	0.428	0.890
	7	0.788	−0.480	2.056	0.429	1.837	0.349	−0.230	0.929	0.523
5	6	0.748	−0.794	2.290	0.522	1.433	0.331	−0.373	1.036	0.784
	7	2.173	0.920	3.426	0.424	5.125	0.962	0.386	1.538	<0.001
6	7	1.425	−0.153	3.002	0.534	2.669	0.631	−0.091	1.353	0.107

), and HEAS-API (

Table A16. Post hoc comparison concerning HEAS-Anxiety for personal impact.

			95% CI for Mean Difference					95% CI for Cohen’s d
		Mean Difference	Lower	Upper	SE	t	Cohen’s d	Lower	Upper	ptukey
1	2	−0.955	−1.529	−0.380	0.195	−4.908	−0.470	−0.764	−0.176	<0.001
	3	0.641	0.060	1.222	0.197	3.258	0.315	0.020	0.611	0.020
	4	2.037	1.193	2.880	0.286	7.132	1.003	0.569	1.437	<0.001
	5	−1.460	−2.279	−0.641	0.277	−5.269	−0.719	−1.138	−0.300	<0.001
	6	0.635	−0.575	1.846	0.410	1.551	0.313	−0.302	0.928	0.714
	7	1.582	0.718	2.447	0.293	5.410	0.779	0.337	1.222	<0.001
2	3	1.595	0.883	2.307	0.241	6.619	0.786	0.420	1.151	<0.001
	4	2.991	2.052	3.930	0.318	9.413	1.473	0.985	1.961	<0.001
	5	−0.505	−1.422	0.411	0.310	−1.629	−0.249	−0.715	0.217	0.663
	6	1.590	0.311	2.868	0.433	3.674	0.783	0.131	1.434	0.005
	7	2.537	1.580	3.494	0.324	7.830	1.249	0.755	1.744	<0.001
3	4	1.396	0.453	2.339	0.319	4.376	0.687	0.206	1.169	<0.001
	5	−2.101	−3.021	−1.180	0.312	−6.743	−1.035	−1.508	−0.561	<0.001
	6	−0.005	−1.287	1.276	0.434	−0.013	−0.003	−0.653	0.648	1.000
	7	0.942	−0.019	1.903	0.325	2.895	0.464	−0.025	0.953	0.059
4	5	−3.497	−4.602	−2.391	0.374	−9.348	−1.722	−2.296	−1.148	<0.001
	6	−1.401	−2.821	0.018	0.481	−2.916	−0.690	−1.413	0.032	0.056
	7	−0.454	−1.594	0.685	0.386	−1.178	−0.224	−0.802	0.355	0.902
5	6	2.095	0.690	3.500	0.476	4.406	1.032	0.315	1.749	<0.001
	7	3.043	1.921	4.164	0.379	8.019	1.498	0.919	2.077	<0.001
6	7	0.947	−0.485	2.380	0.485	1.954	0.467	−0.261	1.194	0.445

). Notably, for all the investigated eco-anxiety factors, the differences between Clusters with a significant differences at p < 0.001 were between Clusters 1 and 2, Clusters 1 and 7, Clusters 2 and 4, and Clusters 5 and 7.

3.3. Multiple Stepwise Regression Analysis: Results

3.3.1. Readiness to Change and Pro-Environmental Behaviors

Concerning RTC and sustainable behaviors, the stepwise multiple linear regression analysis pointed out that the action dimension (RTC) has a significant predictive role for all the pro-environmental behaviors investigated in terms of consumption (Beta = 0.289; t = 8.293; p < 0.001) (

Table A17. Multiple linear regression analysis: Readiness to Change and PEB Conservation.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
RTC-A	0.289	8.293	<0.001	0.281	0.455	0.141
RTC-PR	0.140	4.017	<0.001	0.099	0.289

Notes: RTC, Readiness to Change; A, Actions; PR, Perceived readiness; Sig., Significance; CI, Confidence Interval.

), environmental citizenship (Beta = 0.163; t = 3.875; p < 0.001) (

Table A18. Multiple linear regression analysis: Readiness to Change and PEB Environmental Citizenship.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
RTC-PR	0.152	3.516	<0.001	0.096	0.339	0.155
RTC-A	0.163	3.875	<0.001	0.108	0.329
RTC-M	0.147	3.596	<0.001	0.089	0.304

Notes: RTC, Readiness to Change; PR, Perceived readiness; A, Actions; M, Motivation; Sig., Significance; CI, Confidence Interval.

), food (Beta = 0.217; t = 5.453; p < 0.001) (

Table A19. Multiple linear regression analysis: Readiness to Change and PEB Food.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
RTC-M	0.129	2.411	0.016	0.041	0.397	0.129
RTC-A	0.217	5.453	<0.001	0.240	0.510
RTC-SE	−0.099	−2.803	0.005	−0.253	−0.045
RTC-PI	0.122	2.471	0.014	0.047	0.412

Notes: RTC, Readiness to Change; M, Motivation; A, Action; SE, Self-efficacy; PI, Perceived importance of the problem; Sig., Significance; CI, Confidence Interval.

), and transportation (Beta = 0.158; t = 3.947; p < 0.001) (

Table A20. Multiple linear regression analysis: Readiness to Change and PEB Transportation.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
RTC-PI	0.149	4.077	<0.001	0.081	0.230	0.052
RTC-A	0.158	3.947	<0.001	0.076	0.225
RTC-SE	−0.133	−3.629	<0.001	−0.171	−0.051

Notes: RTC, Readiness to Change; PI, Perceived importance of the problem; A, Action; SE, Self-efficacy; Sig., Significance; CI, Confidence Interval.

).

3.3.2. Readiness to Change and Eco-Anxiety

Regarding the predicting role of RTC on eco-anxiety, the analysis highlighted that the dimensions motivation and perceived importance of the problem had a significant predictive role for all the eco-anxiety factors, except for behavioral symptoms (

Table A21. Multiple linear regression analysis: Readiness to Change and HEAS Affective Symptoms.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
RTC-PI	0.122	3.085	0.002	0.048	0.217	0.073
RTC-ES	−0.134	−3.608	<0.001	−0.175	−0.052
RTC-SE	0.111	2.068	0.039	0.005	0.206
RTC-M	0.122	3.085	0.002	0.048	0.217

Notes: RTC, Readiness to Change; PI, Perceived importance of the problem; ES, Effectiveness of the proposed solution; SE, Self-efficacy; M, Motivation; Sig., Significance; CI, Confidence Interval.

,

Table A22. Multiple linear regression analysis: Readiness to Change and HEAS Rumination.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
RTC-M	0.209	4.167	<0.001	0.076	0.211	0.094
RTC-PI	0.117	2.332	0.020	0.014	0.164

Notes: RTC, Readiness to Change; PI, Perceived importance of the problem; ES, Effectiveness of the proposed solution; SE, Self-efficacy; M, Motivation; Sig., Significance; CI, Confidence Interval.

,

Table A23. Multiple linear regression analysis: Readiness to Change and HEAS Behavioral Symptoms.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
RTC-ES	0.169	4.269	<0.001	0.083	0.224	0.042
RTC-SE	−0.157	−4.019	<0.001	−0.163	−0.056
RTC-A	0.126	3.199	0.001	0.039	0.163

Notes: RTC, Readiness to Change; ES, Effectiveness of the proposed solution; SE, Self-efficacy; A, Action; Sig., Significance; CI, Confidence Interval.

and

Table A24. Multiple linear regression analysis: Readiness to Change and HEAS Anxiety for personal impact.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
RTC-M	0.156	3.010	0.003	0.041	0.194	0.170
RTC-A	0.177	4.734	<0.001	0.079	0.191
RTC-PI	0.174	3.597	<0.001	0.066	0.224
RTC-SS	−0.087	−2.733	0.006	−0.118	−0.019

Notes: RTC, Readiness to Change; M, Motivation; A, Action; PI, Perceived importance of the problem; SS, Social support; Sig., Significance; CI, Confidence Interval.

). Notably, the motivation dimension was significantly and positively associated with eco-anxiety factors in terms of affective symptoms (Beta = 0.122; t = 3.085; p = 0.002) (Table A21), rumination (Beta = 0.209; t = 4.167; p < 0.001) (Table A22), and anxiety for personal impact (Beta = 0.156; t = 3.010; p = 0.003) (Table A24). On the other hand, perceived importance of the problem was significantly and positively associated with affective symptoms (Beta = 0.122; t = 3.085; p = 0.002) (Table A21), rumination (Beta = 0.117; t = 2.332; p = 0.020) (Table A22), and anxiety for personal impact (Beta = 0.174; t = 3.597; p < 0.001) (Table A24).

3.3.3. Sensitivity and Pro-Environmental Behaviors

Regarding sensitivity and pro-environmental behaviors, the stepwise multiple linear regression analysis highlighted that the only HSPS dimension associated with all the considered pro-environmental behaviors was Aesthetic Sensitivity (HSPS) (

Table A25. Multiple linear regression analysis: Sensitivity and PEB Conservation.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
HSPS-AES	0.304	8.021	<0.001	0.198	0.327	0.068
HSPS-EOE	−0.090	−2.381	0.017	−0.072	−0.007

Notes: HSPS, Highly Sensitive Person Scale; AES, Aesthetic Sensitivity; EOE, Ease of Excitation; Sig. Significance; CI, Confidence Interval.

,

Table A26. Multiple linear regression analysis: Sensitivity and PEB Environmental Citizenship.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
HSPS-AES	0.304	8.021	<0.001	0.198	0.327	0.101
HSPS-EOE	−0.090	−2.381	0.017	−0.072	−0.007

Notes: HSPS, Highly Sensitive Person Scale; AES, Aesthetic Sensitivity; EOE, Ease of Excitation; Sig. Significance; CI, Confidence Interval.

,

Table A27. Multiple linear regression analysis: Sensitivity and PEB Food.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
HSPS-AES	0.163	4.661	<0.001	0.110	0.270	0.068
HSPS-LST	0.149	4.269	<0.001	0.077	0.209

Notes: HSPS, Highly Sensitive Person Scale; AES, Aesthetic Sensitivity; LST, Low Sensory Threshold; Sig. Significance; CI, Confidence Interval.

and

Table A28. Multiple linear regression analysis: Sensitivity and PEB Transportation.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
HSPS-AES	0.178	4.706	<0.001	0.068	0.164	0.066
HSPS-EOE	0.115	3.047	0.002	0.014	0.062

Notes: HSPS, Highly Sensitive Person Scale; AES, Aesthetic Sensitivity; EOE, Ease of Excitation; Sig. Significance; CI, Confidence Interval.

). Notably, this kind of dimension was significantly and positively associated with sustainable behaviors in terms of conservation (Beta = 0.304; t = 8.021; p < 0.001) (Table A25), environmental citizenship (Beta = 0.304; t = 8.021; p < 0.001) (Table A26), food (Beta = 0.163; t = 4.661; p < 0.001) (Table A27), and transportation (Beta = 0.178; t = 4.706; p < 0.001) (Table A28).

3.3.4. Sensitivity and Eco-Anxiety

Concerning the predicting role of sensitivity on eco-anxiety, the results pointed out that the HSPS dimensions linked with all the eco-anxiety factors was Low Sensory Threshold (

Table A29. Multiple linear regression analysis: Sensitivity and HEAS Affective symptoms.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
HSPS-EOE	0.308	7.385	<0.001	0.075	0.130	0.200
HSPS-LST	0.134	3.461	0.001	0.031	0.113
HSPS-AES	0.070	1.982	0.048	0.000	0.091

Notes: HSPS, Highly Sensitive Person Scale; EOE, Ease of Excitation; LST, Low Sensory Threshold; AES, Aesthetic Sensitivity; Sig. Significance; CI, Confidence Interval.

,

Table A30. Multiple linear regression analysis: Sensitivity and HEAS Rumination.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
HSPS-AES	0.268	7.903	<0.001	0.094	0.157	0.125
HSPS-LST	0.147	4.341	<0.001	0.031	0.083

Notes: HSPS, Highly Sensitive Person Scale; AES, Aesthetic Sensitivity; LST, Low Sensory Threshold; Sig. Significance; CI, Confidence Interval.

,

Table A31. Multiple linear regression analysis: Sensitivity and HEAS Behavioral symptoms.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
HSPS-LST	0.225	5.595	<0.001	0.065	0.136	0.128
HSPS-EOE	0.171	4.256	<0.001	0.026	0.069

Notes: HSPS, Highly Sensitive Person Scale; LST, Low Sensory Threshold; EOE, Ease of Excitation; Sig. Significance; CI, Confidence Interval.

and

Table A32. Multiple linear regression analysis: Sensitivity and HEAS Anxiety for personal impact.

Variables	Beta	t	Sig.	CI 95%		Adjusted-R2
				Lower Bound	Upper Bound
HSPS-AES	0.273	7.565	<0.001	0.104	0.178	0.164
HSPS-LST	0.108	2.720	0.007	0.013	0.079
HSPS-EOE	0.104	2.430	0.015	0.005	0.049

Notes: HSPS, Highly Sensitive Person Scale; AES, Aesthetic Sensitivity; LST, Low Sensory Threshold; EOE, Ease of Excitation; Sig. Significance; CI, Confidence Interval.

). In detail, the HSPS dimension above was significantly and positively associated with affective symptoms (Beta = 0.134; 3.461; p = 0.001) (Table A29), rumination (Beta = 0.147; t = 4.341; p < 0.001) (Table A30), behavioral symptoms (Beta = 0.225; 5.595; p < 0.001) (Table A31), and anxiety for personal impact (Beta = 0.108; t = 2.720; p = 0.007) (Table A32).

4. Discussion

The present study aimed to observe how sensitivity clusters in relation to Readiness to Change. Furthermore, it sought to investigate potential differences among the identified clusters in terms of engagement in sustainable behaviors and levels of eco-anxiety. In line with this, the number of clusters was fixed at a maximum of seven based on the dimensions of Readiness to Change (RTC), and a RFC (Random Forest Clustering) was conducted.

The results revealed the presence of seven distinct clusters highlighting that sensitivity was the key factor in distinguishing them as well as higher levels of sensitivity are associated with higher RTC values (H1).

Notably, Cluster 1 exhibited the highest dimensionality and explained variance representing people with moderate levels of Readiness to Change, mostly in terms of motivation and perceived readiness, and lower levels of sensitivity. Moreover, together with Cluster 1, and Clusters 2 and 5, representing, respectively, individuals with moderate Readiness to Change and high sensitivity, and people with high Readiness to Change and moderate sensitivity, are characterized by high levels of sustainable behavior engagement and eco-anxiety in terms of rumination and anxiety for personal impact. On the other hand, concerning the affective symptoms of eco-anxiety, Cluster 1, Cluster 5, and Cluster 6 (low Readiness to Change and moderate sensitivity) are characterized by higher mean values in this kind of domain. Finally, Cluster 2, Cluster 3 (low Readiness to Change and sensitivity), and Cluster 5 are represented by higher mean values of behavioral symptoms of eco-anxiety. In general, results pointed out how the combination of Readiness to Change and sensitivity plays a crucial role in the adoption of sustainable behaviors and in experiencing higher levels of eco-anxiety by supporting H1. In fact, it is observed that Cluster 5 (high Readiness to Change and high sensitivity) is the one implicated in all sustainable behaviors and eco-anxiety factors. Furthermore, the Gini Index has highlighted how the ease of excitation component represents the main factor able to differentiate clusters. From our perspective, this could depend on the fact that, among the other dimensions of sensitivity, ease of excitation seems to be the most comprehensive, considering that it represents a phenomenon that derives from both internal and external activations [76]. Finally, despite the importance of sensitivity construct in promoting sustainable behaviors and attitudes, it appears that the construct of Readiness to Change can compensate for low levels of sensitivity, supporting both the adoption of pro-environmental behaviors and higher levels of eco-anxiety.

4.1. Readiness to Change and Sensitivity and Their Role in the Engagement of Sustainable Behaviors and Eco-Anxiety

Regarding the relationship between Readiness to Change and the implementation of pro-environmental behaviors, the existing literature has highlighted a significant link between these phenomena for all the behaviors considered in the present study [27,44]. Accordingly, findings are well supported by the literature, emphasizing the key role of RTC in the adoption of sustainable actions [27,44]. Specifically, in line with previous literature findings [44], the stepwise multiple linear regression analysis pointed out that the action dimension of Readiness to Change is a significant predictor for all types of sustainable behavior considered. In line with the transtheoretical model [37] and the assumptions of Schwarzer et al. [98], the action dimension could play such a significant role considering that it represents the practical implementation of the behavior also in terms of desirability to pursue in behavioral change, including sustainable behaviors [37,44,98].

On the other hand, to date, the relationship between RTC and eco-anxiety levels has been less investigated. Specifically, an association has been observed between readiness to transition to a low-carbon lifestyle and the affective symptomatology of eco-anxiety [23,82,99,100]. However, this kind of link refers to a highly specific behavior by not representing a more general RTC aspect.

Nonetheless, the relationship between Readiness to Change and eco-anxiety can be interpreted on the basis of the intrinsic characteristics of eco-anxiety, particularly in terms of engagement in problem-solving [15,32]. In fact, it has been observed that individuals experiencing eco-anxiety may be more inclined to adopt sustainable behaviors. On this basis, it is potentially reasonable to assume that Readiness to Change plays a crucial role in supporting such engagement [15,44]. In particular, we believe that it is precisely the combination with RTC that makes eco-anxiety a positive and proactive construct, considering that this phenomenon can exacerbate adverse effects in terms of, for example, burnout or paralyzing symptoms [32,101]. In this regard, it is important to underline how eco-anxiety can lead to adverse effects [32,101], and for this reason, from our point of view, interventions, as well as treatments, should be able to support and promote a proactive level of eco-anxiety by hindering the potential emergence of negative consequences that could lead to the person suffering.

Furthermore, the regression analysis highlighted the significant role that the motivation and perceived importance of the problem dimensions of Readiness to Change play in eco-anxiety. This result underlines the complexity of the phenomenon as well as its dual composition [32]. The pro-active form of eco-anxiety seems to be also associated with motivational processes [32]. On the other hand, as regards the role of the perceived importance of the problem, this could depend on the construct of risk perception. In fact, this dimension of Readiness to Change is closely linked to risk perception [44], a process that is also associated with eco-anxiety [102].

Regarding the association between sensitivity and both pro-environmental behaviors and eco-anxiety, the results are well supported by the literature [27]. In fact, Duradoni and colleagues [27] found significant associations between being a highly sensitive person and all the sustainable behaviors examined in the present study, as well as all dimensions of eco-anxiety. The authors suggest that these connections may stem from the characteristics typically associated with highly sensitive individuals [27]. In particular, the stepwise multiple regression analysis pointed out that aesthetic sensitivity played a significant predictive role in all the pro-environmental behaviors considered. This type of association has already been observed in the literature [27], and it may be because this characteristic of sensitivity involves intensely appreciating the beauty of nature [29]. This component could support people in implementing sustainable behaviors.

Specifically, individuals with high sensitivity appear to react more strongly and intensely to environmental stimuli, a type of response that could lead to higher levels of eco-anxiety [27,72,103]. As aforementioned, environmental sensitivity could also be linked with greater engagement in sustainable behaviors due to empathy characteristics of sensitive people toward the natural environment [27,30]: as a result, highly sensitive people seem to be more inclined to take action to protect the environment [27,104,105]. Moreover, the analysis also showed that the Low Sensory Threshold component of sensitivity was significantly linked with all the eco-anxiety factors, which may be due to the fact that this dimension refers to the emergence of sensory reactions in response to external stimulation [76].

In summary, the results highlighted the complexity of the investigated phenomena, as well as the effectiveness of the combination of psychological determinants, such as RTC and sensitivity construct, in promoting PEBs.

Moreover, the results also pointed out that high sensitivity seems to be linked with greater RTC. These findings are in line with previous studies suggesting that this kind of trait may potentially be linked to the Behavioral Activation System [73,75,76], supporting the hypothesis that, although it shares certain characteristics with the Behavioral Inhibition System [106], sensitivity may constitute a distinct and autonomous construct [27].

Taking this perspective into account, in our point of view, these results may contribute to a deeper understanding of the sensitivity trait, particularly within the context of sustainability. Specifically, in Cluster 2, higher levels of sensitivity are especially associated with higher scores in the perceived importance, motivation, effectiveness of the proposed solution, action, and perceived readiness dimensions of RTC.

Regarding the relationship between perceived importance and the sensitivity trait, this kind of link may be explained by existing literature data related to the compassion and empathy characteristics of highly sensitive people towards natural context and environment [30]. Such emotional dispositions may translate into a heightened perception of the importance of addressing climate change in order to protect the environment.

The association between motivation and sensitivity may also stem from the emotional connection that highly sensitive people experience with nature, as well as from the positive emotions they derive from it [27]. Indeed, the literature emphasizes that the motivational power of emotions plays a crucial role in driving behaviors aimed at environmental protection [31].

Concerning the link with the dimension effectiveness of the proposed solution, considering the meaning of this dimension in terms of feeling confident in acting for contrasting climate crisis [44], the association above may be due the strong engagement of sensitivity people in promoting pro-environmental behaviors [27]. Notably, it is presumable that people strongly engaged in sustainable lifestyle also significantly believe in the efficacy of pro-environmental behaviors. However, future studies are needed for a deeper investigation into the interpretation above. Moreover, Aron and Aron [29] point out that the sensitivity trait, understood as sensory processing sensitivity, may activate two distinct processes: exploration or avoidance. The adoption of one strategy over the other depends on how sensory information is transmitted and processed at the neurological level, an aspect widely used in the literature to describe environmental sensitivity [29,72]. Within this theoretical framework, the relationship observed between sensitivity and the action dimension of Readiness to Change may represent one of the underlying mechanisms driving exploratory behaviors.

Finally, it is presumable that the association between sensitivity and perceived readiness may be linked to that fact that sensitive people may feel more ready to be engaged in sustainable behaviors for supporting natural environment and contrasting climate crisis due to their connection with nature [27,76,77]. In conclusion, the sensitivity trait appears to be associated with Readiness to Change, a psychological dimension that, due to its characteristics, may represent a key predisposing factor for the activation of exploratory and proactive behavioral strategies.

4.2. Future Implications

The present work highlighted how Readiness to Change, when combined with high sensitivity, has an impact on the enactment of pro-environmental behaviors as well as on eco-anxiety, which, if interpreted as a factor that promotes problem-solving, could also represent a psychological determinant useful for adopting sustainable lifestyles. Nonetheless, the data also show that Readiness to Change, even when not associated with high sensitivity (Cluster 5), still plays an important role in the implementation of pro-environmental behaviors as well as eco-anxiety. This result, in particular, can provide a solid foundation for the development of interventions that are helpful in addressing the climate crisis. In detail, programs focused on supporting Readiness to Change may also be effective among people without a strong environmental sensitivity. Moreover, as a result, the findings underscore the need to promote both Readiness to Change and sensitivity in order to encourage the adoption of more sustainable lifestyles, aiming to improve the presence of both components, which are beneficial for greater sustainable engagement. In particular, these results could serve as groundwork for the development of future environmental awareness and education programs, as well as policies, to foster Readiness to Change and, consequently, involvement in pro-environmental behaviors. Furthermore, as suggested by Duradoni and colleagues [27] and based on the data regarding sensitivity, educational interventions could focus on strengthening the human–nature connection in order to support a deeper bond with the natural world and promote actions to protect it. Notably, educational programs can support the human–nature connection through the development of nature engagement activities, as well as interventions for restoring natural environments [107]. In particular, from our perspective, this last type of intervention could support the action dimension of Readiness to Change, which, as emerged from the present study, appears to play a key role in the engagement of pro-environmental behaviors. Furthermore, considering the role played by the perceived importance of the problem and motivation, the development of interventions capable of promoting the above-mentioned dimensions is also desirable. In this regard, Sinclair and colleagues [108] have highlighted how interventions such as, for example, guided imagery and action planning are able to work on risk perception which is closely linked to the perceived importance of the problem [44]. On the other hand, programs aimed at fostering reflections on the future concerning the environment appear to play a crucial role in stimulating active motivation [108]. In addition, the aforementioned awareness campaigns could leverage environmental sensitivity to further engage communities in environmental protection efforts [27].

5. Strengths and Limitations

The present work has several limitations. Firstly, the data were collected using a cross-sectional design; therefore, it is not possible to infer any potential cause–effect relationships among the investigated factors. In keeping with this, further longitudinal or experimental studies are needed to deepen the investigation of the phenomena. Moreover, although the anonymity of the data may have reduced the effect of social desirability bias [86], it is still possible that the data were partially influenced by it. Furthermore, the majority of the sample was composed of a higher proportion of women participants, which may have affected the generalizability of the results. Accordingly, future studies with a more balanced sample are needed. Additionally, the study revealed the presence of a low Dunn index. However, as suggested by the literature [109], the Dunn index may be sensitive to noisy points, and moreover, psychological variables such as Readiness to Change may not represent trait factors but somewhat fluctuating ones, supporting the possible overlap between clusters. Finally, it is also important to underline that the sample was primarily composed of women and student which represent two population categories strongly engaged in sustainable actions [110,111,112]. Based on the characteristics of the sample of this study, the results may not be generalizable. Future studies are therefore needed to observe the investigated phenomena within populations with different socio-demographic characteristics. However, to the best of our knowledge, this is the first paper to create clustered profiles related to Readiness to Change and environmental sensitivity, representing a starting point for the development of future studies, as well as awareness and education programs focused on this specific field of research. Nevertheless, further research is needed to explore and consolidate the findings that have emerged.

6. Conclusions

In conclusion, the present work sheds light on the potential different profiles that take into account both Readiness to Change and traits of environmental sensitivity. Additionally, based on the identified clusters, the study also highlighted which profiles are more engaged in the promotion and implementation of sustainable behaviors, ranging from environmental citizenship to the use of sustainable means of transportation. Finally, the findings may serve as a foundation for both future research in this field and the development of educational and awareness programs focused on environmental issues.