Participants will smoke two visually identical cigarettes, one with a charcoal filter and one with a non-charcoal filter, in counterbalanced order. The charcoal filter cigarette will be a Natural American Spirit Sky cigarette; the non-charcoal filter cigarette will be a Natural American Spirit Orange cigarette.

Risk beliefs were captured using an 8-item scale that asked participants to compare each study cigarette to 'regular' cigarettes on eight statements (i.e., "lower in nicotine", "lower in tar", "less addictive", "less likely to cause cancer", "has fewer chemicals", "is healthier", "makes smoking safer", "helps people quit smoking") using a 5-point response scale (1='definitely untrue', 5='definitely true'). Responses were scored dichotomously ("untrue" responses scored as correct; "unsure"/"true" responses scored as incorrect) and summed to create an overall beliefs measure. Thus, responses are possible on a scale of 0-8, with a score of 0 indicating no correct beliefs and 8 indicating the participant was correct on all belief items.

Perceived health risks were assessed using the mean of six items that asked participants to indicate on a 7-point Likert scale (1 = "very low risk", 7 = "very high risk") their risk of developing smoking-related health conditions (i.e., lung cancer, heart disease, stroke, emphysema, respiratory infections, and other cancers) from regular use of each study cigarette.

Subjective ratings were assessed with THE Cigarette Rating Scale, a 100 mm visual analog scale used by the tobacco industry and our laboratory to assess 14 characteristics; higher scores generally indicate more favorable ratings (e.g., taste: 0 = "bad," 100 = "good"). We examined mean scores from three subscales generated from these items to assess domains of product harshness, smoking satisfaction, and positive sensory experience. We also explored a fourth 'cleanliness' subscale created by averaging two exploratory items assessing "chemical vs. tobacco taste" and "dirty vs. clean body feeling" based on qualitative data from a related pilot study. All subscale scores have a range of 0-100, with higher scores generally indicative of more favorable ratings.

Subjective ratings were assessed with the Cigarette Rating Scale, a 100 mm visual analog scale used by the tobacco industry and our laboratory to assess 14 characteristics; higher scores generally indicate more favorable ratings (e.g., taste: 0 = "bad," 100 = "good"). We examined mean scores from three subscales generated from these items to assess domains of product harshness, smoking satisfaction, and positive sensory experience. We also explored a fourth 'cleanliness' subscale created by averaging two exploratory items assessing "chemical vs. tobacco taste" and "dirty vs. clean body feeling" based on qualitative data from a related pilot study. All subscale scores have a range of 0-100, with higher scores generally indicative of more favorable ratings.

Subjective ratings were assessed with the Cigarette Rating Scale, a 100 mm visual analog scale used by the tobacco industry and our laboratory to assess 14 characteristics; higher scores generally indicate more favorable ratings (e.g., taste: 0 = "bad," 100 = "good"). We examined mean scores from three subscales generated from these items to assess domains of product harshness, smoking satisfaction, and positive sensory experience. We also explored a fourth 'cleanliness' subscale created by averaging two exploratory items assessing "chemical vs. tobacco taste" and "dirty vs. clean body feeling" based on qualitative data from a related pilot study. All subscale scores have a range of 0-100, with higher scores generally indicative of more favorable ratings.

Subjective ratings were assessed with the Cigarette Rating Scale, a 100 mm visual analog scale used by the tobacco industry and our laboratory to assess 14 characteristics; higher scores generally indicate more favorable ratings (e.g., taste: 0 = "bad," 100 = "good"). We examined mean scores from three subscales generated from these items to assess domains of product harshness, smoking satisfaction, and positive sensory experience. We also explored a fourth 'cleanliness' subscale created by averaging two exploratory items assessing "chemical vs. tobacco taste" and "dirty vs. clean body feeling" based on qualitative data from a related pilot study. All subscale scores have a range of 0-100, with higher scores generally indicative of more favorable ratings.

CO will be measured in parts per million (ppm) using the Vitalograph BreathCO carbon monoxide monitor (Lenexa, KS) at the onset of the laboratory visit, as well as before and after each cigarette smoked. CO boost - the change in CO values resulting from smoking a cigarette - crudely estimates smoke exposure due to smoking an individual cigarette.

Total of all reporting groups

Assessed using the question, "In the past 7 days, about how many cigarettes have you smoked per day?"

Assessed using the Fagerstrom Test for Cigarette Dependence (FTCD), a 6-item, self-report measure of nicotine dependence. Items are summed to a score ranging from 0-10, with greater scores indicating greater nicotine dependence.

All participants will smoke two visually identical cigarettes varying in filter type (within-subject condition: charcoal and non-charcoal filter), in counterbalanced order. Participants will be randomized to receive cigarettes in one of two pack colors (between-subject-condition: light or dark).

Smoking behaviors were assessed using video-scored measures of smoking topography (i.e., puffing behavior), including number of puffs taken, total puffing duration, and total interpuff interval (time between puffs). Briefly, research staff used a digital timestamp feature in an open-source video editing software to estimate start and end times for individual puffs based on various physical cues (e.g., inhaling, glowing cigarette tip), prioritizing overall puffing behaviors (e.g., multiple occurrences of tip glowing without removing the cigarette from the mouth ["stutter puffs"] were treated as a single puff).

Smoking behaviors were assessed using video-scored measures of smoking topography (i.e., puffing behavior), including number of puffs taken, total puffing duration, and total interpuff interval (time between puffs). Briefly, research staff used a digital timestamp feature in an open-source video editing software to estimate start and end times for individual puffs based on various physical cues (e.g., inhaling, glowing cigarette tip), prioritizing overall puffing behaviors (e.g., multiple occurrences of tip glowing without removing the cigarette from the mouth ["stutter puffs"] were treated as a single puff).

Risk beliefs were captured using an 8-item scale that asked participants to compare each study cigarette to 'regular' cigarettes on eight statements (i.e., "lower in nicotine", "lower in tar", "less addictive", "less likely to cause cancer", "has fewer chemicals", "is healthier", "makes smoking safer", "helps people quit smoking") using a 5-point response scale (1='definitely untrue', 5='definitely true'). Responses were scored dichotomously ("untrue" responses scored as correct; "unsure"/"true" responses scored as incorrect) and summed to create an overall beliefs measure. Thus, responses are possible on a scale of 0-8, with a score of 0 indicating no correct beliefs and 8 indicating the participant was correct on all belief items.

Perceived health risks were assessed using the mean of six items that asked participants to indicate on a 7-point Likert scale (1 = "very low risk", 7 = "very high risk") their risk of developing smoking-related health conditions (i.e., lung cancer, heart disease, stroke, emphysema, respiratory infections, and other cancers) from regular use of each study cigarette.

Subjective ratings were assessed with THE Cigarette Rating Scale, a 100 mm visual analog scale used by the tobacco industry and our laboratory to assess 14 characteristics; higher scores generally indicate more favorable ratings (e.g., taste: 0 = "bad," 100 = "good"). We examined mean scores from three subscales generated from these items to assess domains of product harshness, smoking satisfaction, and positive sensory experience. We also explored a fourth 'cleanliness' subscale created by averaging two exploratory items assessing "chemical vs. tobacco taste" and "dirty vs. clean body feeling" based on qualitative data from a related pilot study. All subscale scores have a range of 0-100, with higher scores generally indicative of more favorable ratings.

Subjective ratings were assessed with the Cigarette Rating Scale, a 100 mm visual analog scale used by the tobacco industry and our laboratory to assess 14 characteristics; higher scores generally indicate more favorable ratings (e.g., taste: 0 = "bad," 100 = "good"). We examined mean scores from three subscales generated from these items to assess domains of product harshness, smoking satisfaction, and positive sensory experience. We also explored a fourth 'cleanliness' subscale created by averaging two exploratory items assessing "chemical vs. tobacco taste" and "dirty vs. clean body feeling" based on qualitative data from a related pilot study. All subscale scores have a range of 0-100, with higher scores generally indicative of more favorable ratings.

Subjective ratings were assessed with the Cigarette Rating Scale, a 100 mm visual analog scale used by the tobacco industry and our laboratory to assess 14 characteristics; higher scores generally indicate more favorable ratings (e.g., taste: 0 = "bad," 100 = "good"). We examined mean scores from three subscales generated from these items to assess domains of product harshness, smoking satisfaction, and positive sensory experience. We also explored a fourth 'cleanliness' subscale created by averaging two exploratory items assessing "chemical vs. tobacco taste" and "dirty vs. clean body feeling" based on qualitative data from a related pilot study. All subscale scores have a range of 0-100, with higher scores generally indicative of more favorable ratings.

Subjective ratings were assessed with the Cigarette Rating Scale, a 100 mm visual analog scale used by the tobacco industry and our laboratory to assess 14 characteristics; higher scores generally indicate more favorable ratings (e.g., taste: 0 = "bad," 100 = "good"). We examined mean scores from three subscales generated from these items to assess domains of product harshness, smoking satisfaction, and positive sensory experience. We also explored a fourth 'cleanliness' subscale created by averaging two exploratory items assessing "chemical vs. tobacco taste" and "dirty vs. clean body feeling" based on qualitative data from a related pilot study. All subscale scores have a range of 0-100, with higher scores generally indicative of more favorable ratings.

CO will be measured in parts per million (ppm) using the Vitalograph BreathCO carbon monoxide monitor (Lenexa, KS) at the onset of the laboratory visit, as well as before and after each cigarette smoked. CO boost - the change in CO values resulting from smoking a cigarette - crudely estimates smoke exposure due to smoking an individual cigarette.

A hypothetical cigarette purchase task, asked how many study cigarettes participants would purchase in a typical day across a range of escalating prices from free up to $1.45 per cigarette, chosen to reflect double the current median price of Natural American Spirit brands in the U.S. However, because 35-40% of participants failed to reach breakpoint (i.e., the price point at which consumption drops to 0), we report observed values for demand index of intensity (number of cigarettes consumed for free) only.

All participants will be smoke two visually identical cigarettes, one with a charcoal filter and one with a non-charcoal filter (within-subject conditions), in counterbalanced order. Participants will be randomized to receive study cigarettes in one of two pack color conditions: light or dark (between-subject factor).